https://ncorwiki.buffalo.edu/api.php?action=feedcontributions&feedformat=atom&user=JihadNCOR Wiki - User contributions [en]2026-04-12T08:59:18ZUser contributionsMediaWiki 1.43.8https://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73486CTS Ontology Workshop 20232023-03-06T20:56:59Z<p>Jihad: /* Registration */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
[[File:Chs23sm.jpg|link=|right|alt=Charleston]]<br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
* Social Determinants of Health<br />
* Mental health<br />
* ChatGPT<br />
<br />
== '''Organizers''' ==<br />
<br />
'''Workshop Co-organizers'''<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
'''CTSOG Co-chairs'''<br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* '''Medical University of South Carolina, the Biomedical Informatics Center''' and:<br />
<br />
[[File:Markiii-logo_160.png|link=https://www.markiiisys.com/|Mark III Systems]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:Nvidia-logo_180.png|link=https://www.nvidia.com/en-us/|NVIDIA]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:TriNetX-Logo-Horizontal-White_180.jpg|link=http://trinetx.com/|TriNetX]]<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
Registration is free.<br />
<br />
However, registration is required for planning purposes. Registration is now closed.<br />
<br />
== '''Agenda: Thursday Feb 23'''==<br />
<br />
'''08:30a''': Breakfast and Networking<br />
<br />
'''09:00a''': Introduction to the meeting<br />
<br />
'''09:15-10:45a''' Working sessions/Presentations<br />
<br />
*'''Barry Smith and Jobst Landgrebe:''' Discussion on AI and Medicine, with a special focus on ChatGPT<br />
<br />
Slides: [https://buffalo.box.com/v/Smith-on-ChatGPT Smith], [https://buffalo.box.com/v/Landgrebe-ChatGPT Landgrebe]<br />
<br />
How are we to explain the peculiar tendency of ChatGPT to throw up what are called [https://buffalo.box.com/v/ChatGPT-Hallucinations ‘hallucinations’]? To answer this question we will draw on our book ''[https://buffalo.app.box.com/v/AI-Without-Fear Why Machines Will Never Rule the World]'', whose core thesis is that an artificial intelligence that could equal or exceed human intelligence—sometimes called "artificial general intelligence" (AGI)—is for mathematical reasons impossible. The argument for this thesis rests on the fact that (1.) human intelligence is a capability of a complex dynamic system—the human brain and central nervous system, and (2.) systems of this sort (like all organic systems) cannot be modelled mathematically in a way that would allow the models to operate inside a computer. We survey on this basis the potential of AI in the future of clinical and translational research. <br />
<br />
'''--break--'''<br />
<br />
'''11:00-12:30p''' Working sessions/Presentations<br />
<br />
*'''Discussion and brief presentations. Topics to include:<br />
:The Clinical and Translational Science Ontology Landscape in 2023<br />
:Opportunities for AI and Ontologies in medicine<br />
:The State of BFO<br />
:Will AI Make Ontologies Redundant?<br />
<br />
'''12:30-01:30p''' Networking and Working Lunch <br />
<br />
'''01:30-3:00p''' Working sessions/Presentations<br />
<br />
*'''Yonghui Wu:''' A large language model for electronic health records.<br />
<br />
[https://buffalo.box.com/v/Yongui-AI-and-EHR Slides]<br />
<br />
Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
'''--break--'''<br />
<br />
'''3:15-5:00p''' Working sessions/Presentations<br />
<br />
*'''Richard Ohrbach and Barry Smith:''' Defining 'Injury' <br />
<br />
[https://buffalo.box.com/v/Ohrbach-Defining-Injury Slides]<br />
<br />
This session is intended as a contribution to the ontology component of a broader investigation of injury and pain. SNOMED-CT defines an injury as a 'disorder resulting from physical damage to the body'. The WHO defines an injury as a 'bodily lesion at the organic level, resulting from acute exposure to energy (mechanical, thermal, electrical, chemical or radiant), in amounts that exceed the threshold of physiological tolerance.' We will explore these and other definitions with a view to establishing a more coherent understanding of the ontology of injury and of related phenomena such as lesion, trauma, pain, and so forth. Further topics for possible discussion are listed <u>[[Injury | here]]</u>. For the ontology of pain, see in particular [https://philpapers.org/archive/SMITAO-12.pdf here]<br />
<br />
'''--break--'''<br />
<br />
'''5:30p''' Networking Reception<br />
<br />
=='''Agenda: Friday Feb 24'''==<br />
<br />
'''08:30-09:15a''' Breakfast and Networking<br />
<br />
'''09:15-10:15a''' Working sessions/Presentations <br />
<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health<br />
<br />
[https://buffalo.box.com/v/Hogan-Occupations-Ontology Slides]<br />
<br />
The ontological representation of occupation presents several challenges. For example, is it your occupation if you have not yet done it? Are you a nuclear engineer before you start your first job doing it? Also, if an occupation is a role (in the BFO sense), then how does it differ from a job role? If I have an occupation of professor, and have had four job roles as professors at different organizations, then that is one occupation numerically but four job roles. How to handle non-employment situations such as "retired" and "homemaker"? What if you concurrently hold two jobs in different "occupations"? I will discuss these issues in the context of recent discussions at the OMRSE monthly meeting and present some options for ontological representation of occupation and related entities.<br />
<br />
'''--break--'''<br />
<br />
'''10:30-12:30p''' Working sessions/Presentations <br />
<br />
*'''Justin Reese:''' Unsupervised machine learning to define subtypes of long COVID using the Human Phenotype Ontology] [[Authors | Co-authors]] <br />
<br />
[https://docs.google.com/presentation/d/1WKvoNl6yqiA00zH5qHkRfDKTksboTX897zU7yJpT8yY/edit#slide=id.g1dd10f2c7e8_0_4793 Slides]<br />
<br />
Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterized by a wide range of manifestations that are difficult to analyze computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.<br />
<br />
We present a method for defining subtypes of long COVID by computationally modeling PASC phenotype data and applying unsupervised machine learning. We extracted Long COVID phenotype data expressed as HPO (Human Phenotype Ontology) terms from electronic healthcare records (EHRs), and then used semantic similarity of phenotype data to calculate a matrix of pairwise patient similarity. This matrix was then used to clustered patients using unsupervised machine learning (k means clustering).<br />
<br />
We identified six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centers to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems.<br />
<br />
Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.<br />
<br />
*'''Paul Heider:''' Beyond Structure: Using Standardized Labels from a Clinical Section Ontology in Natural Language Processing.<br />
<br />
[https://buffalo.box.com/v/Cllinical-Section-Ontology Slides]<br />
<br />
The structured subdivision of clinical notes into sections is often treated as the primary and ultimate use for automatically tagging section boundaries. We discuss how section information (especially when normalized to a standard ontology) can be used to augment natural language processing algorithms, simplify information retrieval tasks,and speed up system development through disaggregated performance evaluation. Our examples emphasize a Clinical Section Ontology recently developed at MUSC but are applicable to any standardized section ontology. We also include a brief overview of the annotated datasets in this domain.<br />
<br />
'''12:30-01:30p''' Networking and Working Lunch <br />
<br />
'''1:00-03:15p''' Working sessions/Presentations from industry partners<br />
<br />
*'''Andy Lin: (VP Strategy, CTO | Mark III Systems)''' NVIDIA DGX H100 – The Premier AI Platform for Research and LLMs.<br />
<br />
[https://buffalo.box.com/v/Lin-NVIDIA-DGX-H100 Slides]<br />
<br />
With up to 4x training and up to 30x inferencing performance improvements for Large Language Models (LLMs) vs the already industry leading A100 GPU, the recently launched H100 GPU (Hopper) and its purpose-built DGX H100 AI platform are transforming what’s possible with Research and LLMs. In this session, we’ll dive into the DGX H100 platform and how you can get started or accelerate your work with NVIDIA and Mark III.<br />
<br />
*'''John Doole: (TriNetX, LLC)''' Medication Informatics at TriNetX: Powering Clinical Research with a Federated Real World Data Network. <br />
<br />
[https://buffalo.box.com/v/Charleston-TriNetX Slides]<br />
<br />
In this talk we will highlight the importance of medication informatics in clinical research, particularly in the context of a federated network. The talk discusses the challenges of data standardization and interoperability and the benefits of using medication ontologies and classification systems to improve data quality and efficiency. Additionally, the talk presents TriNetX's medication framework and how natural language processing is used to extract medication details, showcasing examples of its implementation and a discussion of the advantages and disadvantages of using NLP in clinical research.<br />
<br />
'''03:15-3:30p''' Closing Remarks<br />
<br />
== '''Participants''' ==<br />
<br />
* Alekseyenko, Alex, Medical University of South Carolina<br />
* Baker, Hamilton, Medical University of South Carolina<br />
* Chatterjee, Prosenjit, The Citadel<br />
* Doole, John, TriNetX<br />
* Heider, Paul, Medical University of South Carolina<br />
* Landgrebe, Jobst, Cognotekt, Cologne<br />
* Hogan, William, co-organizer, University of Florida<br />
* Hutchinson, Tom, University of Pennsylvania, Institute for BioInformatics<br />
* Kim, Wonhee, Visiting Professor, Medical University of South Carolina<br />
* Masci, Anna Maria, National Institute of Environmental Health Sciences (NIEHS)<br />
* Obeid, Jihad, co-organizer, Medical University of South Carolina<br />
* Ohrbach, Richard, University at Buffalo<br />
* Reese, Justin, Lawrence Berkeley National Laboratory<br />
* Scheuermann, Richard, J. Craig Venter Institute<br />
* Simpson, Kit, Medical University of South Carolina<br />
* Skowronek, Matt, TriNetX<br />
* Smith, Barry, co-organizer, University at Buffalo<br />
* Topaloglu, Umit, Center for Biomedical Informatics and Information Technology (CBIIT) National Cancer Institute<br />
* Wehbe, Ramsey, Medical University of South Carolina<br />
* Yonghui Wu, University of Florida<br />
* Zheng, Jie, University of Pennsylvania</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73446CTS Ontology Workshop 20232023-02-24T14:08:45Z<p>Jihad: /* Participants */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
[[File:Chs23sm.jpg|link=|right|alt=Charleston]]<br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
* Social Determinants of Health<br />
* Mental health<br />
* ChatGPT<br />
<br />
== '''Organizers''' ==<br />
<br />
'''Workshop Co-organizers'''<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
'''CTSOG Co-chairs'''<br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* '''Medical University of South Carolina, the Biomedical Informatics Center''' and:<br />
<br />
[[File:Markiii-logo_160.png|link=https://www.markiiisys.com/|Mark III Systems]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:Nvidia-logo_180.png|link=https://www.nvidia.com/en-us/|NVIDIA]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:TriNetX-Logo-Horizontal-White_180.jpg|link=http://trinetx.com/|TriNetX]]<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Agenda: Thursday Feb 23'''==<br />
<br />
'''08:30a''': Breakfast and Networking<br />
<br />
'''09:00a''': Introduction to the meeting<br />
<br />
'''09:15-10:45a''' Working sessions/Presentations<br />
<br />
*'''Barry Smith and Jobst Landgrebe:''' Discussion on AI and Medicine, with a special focus on ChatGPT<br />
<br />
How are we to explain the peculiar tendency of ChatGPT to throw up what are called [https://buffalo.box.com/v/ChatGPT-Hallucinations ‘hallucinations’]? To answer this question we will draw on our book ''[https://buffalo.app.box.com/v/AI-Without-Fear Why Machines Will Never Rule the World]'', whose core thesis is that an artificial intelligence that could equal or exceed human intelligence—sometimes called "artificial general intelligence" (AGI)—is for mathematical reasons impossible. The argument for this thesis rests on the fact that (1.) human intelligence is a capability of a complex dynamic system—the human brain and central nervous system, and (2.) systems of this sort (like all organic systems) cannot be modelled mathematically in a way that would allow the models to operate inside a computer. We survey on this basis the potential of AI in the future of clinical and translational research. <br />
<br />
'''--break--'''<br />
<br />
'''11:00-12:30p''' Working sessions/Presentations<br />
<br />
*'''Discussion and brief presentations. Topics to include:<br />
:The Clinical and Translational Science Ontology Landscape in 2023<br />
:Opportunities for AI and Ontologies in medicine<br />
:The State of BFO<br />
:Will AI Make Ontologies Redundant?<br />
<br />
'''12:30-01:30p''' Networking and Working Lunch <br />
<br />
'''01:30-3:00p''' Working sessions/Presentations<br />
<br />
*'''Yonghui Wu:''' A large language model for electronic health records.<br />
<br />
Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
'''--break--'''<br />
<br />
'''3:15-5:00p''' Working sessions/Presentations<br />
<br />
*'''Richard Ohrbach and Barry Smith:''' Defining 'Injury'<br />
<br />
This session is intended as a contribution to the ontology component of a broader investigation of injury and pain. SNOMED-CT defines an injury as a 'disorder resulting from physical damage to the body'. The WHO defines an injury as a 'bodily lesion at the organic level, resulting from acute exposure to energy (mechanical, thermal, electrical, chemical or radiant), in amounts that exceed the threshold of physiological tolerance.' We will explore these and other definitions with a view to establishing a more coherent understanding of the ontology of injury and of related phenomena such as lesion, trauma, pain, and so forth. Further topics for possible discussion are listed <u>[[Injury | here]]</u>. For the ontology of pain, see in particular [https://philpapers.org/archive/SMITAO-12.pdf here]<br />
<br />
'''--break--'''<br />
<br />
'''5:30p''' Networking Reception<br />
<br />
=='''Agenda: Friday Feb 24'''==<br />
<br />
'''08:30-09:15a''' Breakfast and Networking<br />
<br />
'''09:15-10:15a''' Working sessions/Presentations <br />
<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health.<br />
<br />
The ontological representation of occupation presents several challenges. For example, is it your occupation if you have not yet done it? Are you a nuclear engineer before you start your first job doing it? Also, if an occupation is a role (in the BFO sense), then how does it differ from a job role? If I have an occupation of professor, and have had four job roles as professors at different organizations, then that is one occupation numerically but four job roles. How to handle non-employment situations such as "retired" and "homemaker"? What if you concurrently hold two jobs in different "occupations"? I will discuss these issues in the context of recent discussions at the OMRSE monthly meeting and present some options for ontological representation of occupation and related entities.<br />
<br />
'''--break--'''<br />
<br />
'''10:30-12:30p''' Working sessions/Presentations <br />
<br />
*'''Justin Reese:''' Unsupervised machine learning to define subtypes of long COVID using the Human Phenotype Ontology. [[Authors | Co-authors]] <br />
<br />
Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterized by a wide range of manifestations that are difficult to analyze computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.<br />
<br />
We present a method for defining subtypes of long COVID by computationally modeling PASC phenotype data and applying unsupervised machine learning. We extracted Long COVID phenotype data expressed as HPO (Human Phenotype Ontology) terms from electronic healthcare records (EHRs), and then used semantic similarity of phenotype data to calculate a matrix of pairwise patient similarity. This matrix was then used to clustered patients using unsupervised machine learning (k means clustering).<br />
<br />
We identified six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centers to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems.<br />
<br />
Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.<br />
<br />
*'''Paul Heider:''' Beyond Structure: Using Standardized Labels from a Clinical Section Ontology in Natural Language Processing.<br />
<br />
The structured subdivision of clinical notes into sections is often treated as the primary and ultimate use for automatically tagging section boundaries. We discuss how section information (especially when normalized to a standard ontology) can be used to augment natural language processing algorithms, simplify information retrieval tasks,and speed up system development through disaggregated performance evaluation. Our examples emphasize a Clinical Section Ontology recently developed at MUSC but are applicable to any standardized section ontology. We also include a brief overview of the annotated datasets in this domain.<br />
<br />
'''12:30-01:30p''' Networking and Working Lunch <br />
<br />
'''1:00-03:15p''' Working sessions/Presentations from industry partners<br />
<br />
*'''Andy Lin: (VP Strategy, CTO | Mark III Systems)''' NVIDIA DGX H100 – The Premier AI Platform for Research and LLMs.<br />
<br />
With up to 4x training and up to 30x inferencing performance improvements for Large Language Models (LLMs) vs the already industry leading A100 GPU, the recently launched H100 GPU (Hopper) and its purpose-built DGX H100 AI platform are transforming what’s possible with Research and LLMs. In this session, we’ll dive into the DGX H100 platform and how you can get started or accelerate your work with NVIDIA and Mark III.<br />
<br />
*'''John Doole: (TriNetX, LLC)''' Medication Informatics at TriNetX: Powering Clinical Research with a Federated Real World Data Network.<br />
<br />
In this talk we will highlight the importance of medication informatics in clinical research, particularly in the context of a federated network. The talk discusses the challenges of data standardization and interoperability and the benefits of using medication ontologies and classification systems to improve data quality and efficiency. Additionally, the talk presents TriNetX's medication framework and how natural language processing is used to extract medication details, showcasing examples of its implementation and a discussion of the advantages and disadvantages of using NLP in clinical research.<br />
<br />
'''03:15-3:30p''' Closing Remarks<br />
<br />
== '''Participants''' ==<br />
<br />
* Alekseyenko, Alex, Medical University of South Carolina<br />
* Baker, Hamilton, Medical University of South Carolina<br />
* Chatterjee, Prosenjit, The Citadel<br />
* Doole, John, TriNetX<br />
* Heider, Paul, Medical University of South Carolina<br />
* Landgrebe, Jobst, Cognotekt, Cologne<br />
* Hogan, William, co-organizer, University of Florida<br />
* Hutchinson, Tom, University of Pennsylvania, Institute for BioInformatics<br />
* Kim, Wonhee, Visiting Professor, Medical University of South Carolina<br />
* Masci, Anna Maria, National Institute of Environmental Health Sciences (NIEHS)<br />
* Obeid, Jihad, co-organizer, Medical University of South Carolina<br />
* Ohrbach, Richard, University at Buffalo<br />
* Reese, Justin, Lawrence Berkeley National Laboratory<br />
* Scheuermann, Richard, J. Craig Venter Institute<br />
* Simpson, Kit, Medical University of South Carolina<br />
* Skowronek, Matt, TriNetX<br />
* Smith, Barry, co-organizer, University at Buffalo<br />
* Topaloglu, Umit, Center for Biomedical Informatics and Information Technology (CBIIT) National Cancer Institute<br />
* Wehbe, Ramsey, Medical University of South Carolina<br />
* Yonghui Wu, University of Florida<br />
* Zheng, Jie, University of Pennsylvania</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73423CTS Ontology Workshop 20232023-02-23T18:21:05Z<p>Jihad: /* Agenda: Thursday Feb 23 */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
[[File:Chs23sm.jpg|link=|right|alt=Charleston]]<br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
* Social Determinants of Health<br />
* Mental health<br />
* ChatGPT<br />
<br />
== '''Organizers''' ==<br />
<br />
'''Workshop Co-organizers'''<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
'''CTSOG Co-chairs'''<br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* '''Medical University of South Carolina, the Biomedical Informatics Center''' and:<br />
<br />
[[File:Markiii-logo_160.png|link=https://www.markiiisys.com/|Mark III Systems]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:Nvidia-logo_180.png|link=https://www.nvidia.com/en-us/|NVIDIA]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:TriNetX-Logo-Horizontal-White_180.jpg|link=http://trinetx.com/|TriNetX]]<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Agenda: Thursday Feb 23'''==<br />
<br />
'''08:30a''': Breakfast and Networking<br />
<br />
'''09:00a''': Introduction to the meeting<br />
<br />
'''09:15-10:45a''' Working sessions/Presentations<br />
<br />
*'''Barry Smith and Jobst Landgrebe:''' Discussion on AI and Medicine, with a special focus on ChatGPT<br />
<br />
How are we to explain the peculiar tendency of ChatGPT to throw up what are called [https://buffalo.box.com/v/ChatGPT-Hallucinations ‘hallucinations’]? To answer this question we will draw on our book ''[https://buffalo.app.box.com/v/AI-Without-Fear Why Machines Will Never Rule the World]'', whose core thesis is that an artificial intelligence that could equal or exceed human intelligence—sometimes called "artificial general intelligence" (AGI)—is for mathematical reasons impossible. The argument for this thesis rests on the fact that (1.) human intelligence is a capability of a complex dynamic system—the human brain and central nervous system, and (2.) systems of this sort (like all organic systems) cannot be modelled mathematically in a way that would allow the models to operate inside a computer. We survey on this basis the potential of AI in the future of clinical and translational research. <br />
<br />
'''--break--'''<br />
<br />
'''11:00-12:30p''' Working sessions/Presentations<br />
<br />
*'''Discussion and brief presentations. Topics to include:<br />
:The Clinical and Translational Science Ontology Landscape in 2023<br />
:Opportunities for AI and Ontologies in medicine<br />
:The State of BFO<br />
:Will AI Make Ontologies Redundant?<br />
<br />
'''12:30-01:30p''' Networking and Working Lunch <br />
<br />
'''01:30-3:00p''' Working sessions/Presentations<br />
<br />
*'''Yonghui Wu:''' A large language model for electronic health records.<br />
<br />
Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
'''--break--'''<br />
<br />
'''3:15-5:00p''' Working sessions/Presentations<br />
<br />
*'''Richard Ohrbach and Barry Smith:''' Defining 'Injury'<br />
<br />
This session is intended as a contribution to the ontology component of a broader investigation of injury and pain. SNOMED-CT defines an injury as a 'disorder resulting from physical damage to the body'. The WHO defines an injury as a 'bodily lesion at the organic level, resulting from acute exposure to energy (mechanical, thermal, electrical, chemical or radiant), in amounts that exceed the threshold of physiological tolerance.' We will explore these and other definitions with a view to establishing a more coherent understanding of the ontology of injury and of related phenomena such as lesion, trauma, pain, and so forth. Further topics for possible discussion are listed <u>[[Injury | here]]</u>. For the ontology of pain, see in particular [https://philpapers.org/archive/SMITAO-12.pdf here]<br />
<br />
'''--break--'''<br />
<br />
'''5:30p''' Networking Reception<br />
<br />
=='''Agenda: Friday Feb 24'''==<br />
<br />
'''08:30-09:15a''' Breakfast and Networking<br />
<br />
'''09:15-10:15a''' Working sessions/Presentations <br />
<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health.<br />
<br />
The ontological representation of occupation presents several challenges. For example, is it your occupation if you have not yet done it? Are you a nuclear engineer before you start your first job doing it? Also, if an occupation is a role (in the BFO sense), then how does it differ from a job role? If I have an occupation of professor, and have had four job roles as professors at different organizations, then that is one occupation numerically but four job roles. How to handle non-employment situations such as "retired" and "homemaker"? What if you concurrently hold two jobs in different "occupations"? I will discuss these issues in the context of recent discussions at the OMRSE monthly meeting and present some options for ontological representation of occupation and related entities.<br />
<br />
'''--break--'''<br />
<br />
'''10:30-12:30p''' Working sessions/Presentations <br />
<br />
*'''Justin Reese:''' Unsupervised machine learning to define subtypes of long COVID using the Human Phenotype Ontology. [[Authors | Co-authors]] <br />
<br />
Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterized by a wide range of manifestations that are difficult to analyze computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.<br />
<br />
We present a method for defining subtypes of long COVID by computationally modeling PASC phenotype data and applying unsupervised machine learning. We extracted Long COVID phenotype data expressed as HPO (Human Phenotype Ontology) terms from electronic healthcare records (EHRs), and then used semantic similarity of phenotype data to calculate a matrix of pairwise patient similarity. This matrix was then used to clustered patients using unsupervised machine learning (k means clustering).<br />
<br />
We identified six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centers to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems.<br />
<br />
Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.<br />
<br />
*'''Paul Heider:''' Beyond Structure: Using Standardized Labels from a Clinical Section Ontology in Natural Language Processing.<br />
<br />
The structured subdivision of clinical notes into sections is often treated as the primary and ultimate use for automatically tagging section boundaries. We discuss how section information (especially when normalized to a standard ontology) can be used to augment natural language processing algorithms, simplify information retrieval tasks,and speed up system development through disaggregated performance evaluation. Our examples emphasize a Clinical Section Ontology recently developed at MUSC but are applicable to any standardized section ontology. We also include a brief overview of the annotated datasets in this domain.<br />
<br />
'''12:30-01:30p''' Networking and Working Lunch <br />
<br />
'''1:00-03:15p''' Working sessions/Presentations from industry partners<br />
<br />
*'''Andy Lin: (VP Strategy, CTO | Mark III Systems)''' NVIDIA DGX H100 – The Premier AI Platform for Research and LLMs.<br />
<br />
With up to 4x training and up to 30x inferencing performance improvements for Large Language Models (LLMs) vs the already industry leading A100 GPU, the recently launched H100 GPU (Hopper) and its purpose-built DGX H100 AI platform are transforming what’s possible with Research and LLMs. In this session, we’ll dive into the DGX H100 platform and how you can get started or accelerate your work with NVIDIA and Mark III.<br />
<br />
*'''John Doole: (TriNetX, LLC)''' Medication Informatics at TriNetX: Powering Clinical Research with a Federated Real World Data Network.<br />
<br />
In this talk we will highlight the importance of medication informatics in clinical research, particularly in the context of a federated network. The talk discusses the challenges of data standardization and interoperability and the benefits of using medication ontologies and classification systems to improve data quality and efficiency. Additionally, the talk presents TriNetX's medication framework and how natural language processing is used to extract medication details, showcasing examples of its implementation and a discussion of the advantages and disadvantages of using NLP in clinical research.<br />
<br />
'''03:15-3:30p''' Closing Remarks<br />
<br />
== '''Participants''' ==<br />
<br />
* Alekseyenko, Alex, Medical University of South Carolina<br />
* Baker, Hamilton, Medical University of South Carolina<br />
* Doole, John, TriNetX<br />
* Heider, Paul, Medical University of South Carolina<br />
* Landgrebe, Jobst, Cognotekt, Cologne<br />
* Hogan, William, co-organizer, University of Florida<br />
* Hutchinson, Tom, University of Pennsylvania, Institute for BioInformatics<br />
* Kim, Wonhee, Visiting Professor, Medical University of South Carolina<br />
* Masci, Anna Maria, National Institute of Environmental Health Sciences (NIEHS)<br />
* Obeid, Jihad, co-organizer, Medical University of South Carolina<br />
* Ohrbach, Richard, University at Buffalo<br />
* Reese, Justin, Lawrence Berkeley National Laboratory<br />
* Scheuermann, Richard, J. Craig Venter Institute<br />
* Simpson, Kit, Medical University of South Carolina<br />
* Skowronek, Matt, TriNetX<br />
* Smith, Barry, co-organizer, University at Buffalo<br />
* Topaloglu, Umit, Center for Biomedical Informatics and Information Technology (CBIIT) National Cancer Institute<br />
* Wehbe, Ramsey, Medical University of South Carolina<br />
* Yonghui Wu, University of Florida<br />
* Chatterjee, Prosenjit, The Citadel<br />
* Zheng, Jie, University of Pennsylvania</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73422CTS Ontology Workshop 20232023-02-23T14:26:07Z<p>Jihad: /* Participants */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
[[File:Chs23sm.jpg|link=|right|alt=Charleston]]<br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
* Social Determinants of Health<br />
* Mental health<br />
* ChatGPT<br />
<br />
== '''Organizers''' ==<br />
<br />
'''Workshop Co-organizers'''<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
'''CTSOG Co-chairs'''<br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* '''Medical University of South Carolina, the Biomedical Informatics Center''' and:<br />
<br />
[[File:Markiii-logo_160.png|link=https://www.markiiisys.com/|Mark III Systems]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:Nvidia-logo_180.png|link=https://www.nvidia.com/en-us/|NVIDIA]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:TriNetX-Logo-Horizontal-White_180.jpg|link=http://trinetx.com/|TriNetX]]<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Agenda: Thursday Feb 23'''==<br />
<br />
'''08:30a''': Breakfast and Networking<br />
<br />
'''09:00a''': Introduction to the meeting<br />
<br />
'''09:15-10:45a''' Working sessions/Presentations<br />
<br />
*'''Barry Smith and Jobst Landgrebe:''' Discussion on AI and Medicine, with a special focus on ChatGPT<br />
<br />
How are we to explain the peculiar tendency of ChatGPT to throw up what are called [https://buffalo.box.com/v/ChatGPT-Hallucinations ‘hallucinations’]? To answer this question we will draw on our book ''[https://buffalo.app.box.com/v/AI-Without-Fear Why Machines Will Never Rule the World]'', whose core thesis is that an artificial intelligence that could equal or exceed human intelligence—sometimes called "artificial general intelligence" (AGI)—is for mathematical reasons impossible. The argument for this thesis rests on the fact that (1.) human intelligence is a capability of a complex dynamic system—the human brain and central nervous system, and (2.) systems of this sort (like all organic systems) cannot be modelled mathematically in a way that would allow the models to operate inside a computer. We survey on this basis the potential of AI in the future of clinical and translational research. <br />
<br />
'''--break--'''<br />
<br />
'''11:00-12:30p''' Working sessions/Presentations<br />
<br />
*'''Discussion and brief presentations. Topics to include:<br />
:The Clinical and Translational Science ontology landscape in 2023<br />
:Opportunities for AI in medicine<br />
:Will AI make ontologies redundant?<br />
:TBD<br />
<br />
'''12:30-01:30p''' Networking and Working Lunch <br />
<br />
'''01:30-3:00p''' Working sessions/Presentations<br />
<br />
*'''Yonghui Wu:''' A large language model for electronic health records.<br />
<br />
Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
'''--break--'''<br />
<br />
'''3:15-5:00p''' Working sessions/Presentations<br />
<br />
*'''Richard Ohrbach and Barry Smith:''' Defining 'Injury'<br />
<br />
This session is intended as a contribution to the ontology component of a broader investigation of injury and pain. SNOMED-CT defines an injury as a 'disorder resulting from physical damage to the body'. The WHO defines an injury as a 'bodily lesion at the organic level, resulting from acute exposure to energy (mechanical, thermal, electrical, chemical or radiant), in amounts that exceed the threshold of physiological tolerance.' We will explore these and other definitions with a view to establishing a more coherent understanding of the ontology of injury and of related phenomena such as lesion, trauma, pain, and so forth. Further topics for possible discussion are listed <u>[[Injury | here]]</u>. For the ontology of pain, see in particular [https://philpapers.org/archive/SMITAO-12.pdf here]<br />
<br />
'''--break--'''<br />
<br />
'''5:30p''' Networking Reception<br />
<br />
=='''Agenda: Friday Feb 24'''==<br />
<br />
'''08:30-09:15a''' Breakfast and Networking<br />
<br />
'''09:15-10:15a''' Working sessions/Presentations <br />
<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health.<br />
<br />
The ontological representation of occupation presents several challenges. For example, is it your occupation if you have not yet done it? Are you a nuclear engineer before you start your first job doing it? Also, if an occupation is a role (in the BFO sense), then how does it differ from a job role? If I have an occupation of professor, and have had four job roles as professors at different organizations, then that is one occupation numerically but four job roles. How to handle non-employment situations such as "retired" and "homemaker"? What if you concurrently hold two jobs in different "occupations"? I will discuss these issues in the context of recent discussions at the OMRSE monthly meeting and present some options for ontological representation of occupation and related entities.<br />
<br />
'''--break--'''<br />
<br />
'''10:30-12:30p''' Working sessions/Presentations <br />
<br />
*'''Justin Reese:''' Unsupervised machine learning to define subtypes of long COVID using the Human Phenotype Ontology. [[Authors | Co-authors]] <br />
<br />
Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterized by a wide range of manifestations that are difficult to analyze computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.<br />
<br />
We present a method for defining subtypes of long COVID by computationally modeling PASC phenotype data and applying unsupervised machine learning. We extracted Long COVID phenotype data expressed as HPO (Human Phenotype Ontology) terms from electronic healthcare records (EHRs), and then used semantic similarity of phenotype data to calculate a matrix of pairwise patient similarity. This matrix was then used to clustered patients using unsupervised machine learning (k means clustering).<br />
<br />
We identified six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centers to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems.<br />
<br />
Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.<br />
<br />
*'''Paul Heider:''' Beyond Structure: Using Standardized Labels from a Clinical Section Ontology in Natural Language Processing.<br />
<br />
The structured subdivision of clinical notes into sections is often treated as the primary and ultimate use for automatically tagging section boundaries. We discuss how section information (especially when normalized to a standard ontology) can be used to augment natural language processing algorithms, simplify information retrieval tasks,and speed up system development through disaggregated performance evaluation. Our examples emphasize a Clinical Section Ontology recently developed at MUSC but are applicable to any standardized section ontology. We also include a brief overview of the annotated datasets in this domain.<br />
<br />
'''12:30-01:30p''' Networking and Working Lunch <br />
<br />
'''1:00-03:15p''' Working sessions/Presentations from industry partners<br />
<br />
*'''Andy Lin: (VP Strategy, CTO | Mark III Systems)''' NVIDIA DGX H100 – The Premier AI Platform for Research and LLMs.<br />
<br />
With up to 4x training and up to 30x inferencing performance improvements for Large Language Models (LLMs) vs the already industry leading A100 GPU, the recently launched H100 GPU (Hopper) and its purpose-built DGX H100 AI platform are transforming what’s possible with Research and LLMs. In this session, we’ll dive into the DGX H100 platform and how you can get started or accelerate your work with NVIDIA and Mark III.<br />
<br />
*'''John Doole: (TriNetX, LLC)''' Medication Informatics at TriNetX: Powering Clinical Research with a Federated Real World Data Network.<br />
<br />
In this talk we will highlight the importance of medication informatics in clinical research, particularly in the context of a federated network. The talk discusses the challenges of data standardization and interoperability and the benefits of using medication ontologies and classification systems to improve data quality and efficiency. Additionally, the talk presents TriNetX's medication framework and how natural language processing is used to extract medication details, showcasing examples of its implementation and a discussion of the advantages and disadvantages of using NLP in clinical research.<br />
<br />
'''03:15-3:30p''' Closing Remarks<br />
<br />
== '''Participants''' ==<br />
<br />
* Alekseyenko, Alex, Medical University of South Carolina<br />
* Baker, Hamilton, Medical University of South Carolina<br />
* Doole, John, TriNetX<br />
* Heider, Paul, Medical University of South Carolina<br />
* Landgrebe, Jobst, Cognotekt, Cologne<br />
* Hogan, William, co-organizer, University of Florida<br />
* Hutchinson, Tom, University of Pennsylvania, Institute for BioInformatics<br />
* Kim, Wonhee, Visiting Professor, Medical University of South Carolina<br />
* Masci, Anna Maria, National Institute of Environmental Health Sciences (NIEHS)<br />
* Obeid, Jihad, co-organizer, Medical University of South Carolina<br />
* Ohrbach, Richard, University at Buffalo<br />
* Reese, Justin, Lawrence Berkeley National Laboratory<br />
* Scheuermann, Richard, J. Craig Venter Institute<br />
* Simpson, Kit, Medical University of South Carolina<br />
* Skowronek, Matt, TriNetX<br />
* Smith, Barry, co-organizer, University at Buffalo<br />
* Topaloglu, Umit, Center for Biomedical Informatics and Information Technology (CBIIT) National Cancer Institute<br />
* Wehbe, Ramsey, Medical University of South Carolina<br />
* Yonghui Wu, University of Florida<br />
* Chatterjee, Prosenjit, The Citadel<br />
* Zheng, Jie, University of Pennsylvania</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73421CTS Ontology Workshop 20232023-02-22T19:48:30Z<p>Jihad: /* Participants */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
[[File:Chs23sm.jpg|link=|right|alt=Charleston]]<br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
* Social Determinants of Health<br />
* Mental health<br />
* ChatGPT<br />
<br />
== '''Organizers''' ==<br />
<br />
'''Workshop Co-organizers'''<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
'''CTSOG Co-chairs'''<br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* '''Medical University of South Carolina, the Biomedical Informatics Center''' and:<br />
<br />
[[File:Markiii-logo_160.png|link=https://www.markiiisys.com/|Mark III Systems]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:Nvidia-logo_180.png|link=https://www.nvidia.com/en-us/|NVIDIA]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:TriNetX-Logo-Horizontal-White_180.jpg|link=http://trinetx.com/|TriNetX]]<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Agenda: Thursday Feb 23'''==<br />
<br />
'''08:30a''': Breakfast and Networking<br />
<br />
'''09:00a''': Introduction to the meeting<br />
<br />
'''09:15-10:45a''' Working sessions/Presentations<br />
<br />
*'''Barry Smith and Jobst Landgrebe:''' Discussion on AI and Medicine, with a special focus on ChatGPT<br />
<br />
How are we to explain the peculiar tendency of ChatGPT to throw up what are called [https://buffalo.box.com/v/ChatGPT-Hallucinations ‘hallucinations’]? To answer this question we will draw on our book ''[https://buffalo.app.box.com/v/AI-Without-Fear Why Machines Will Never Rule the World]'', whose core thesis is that an artificial intelligence that could equal or exceed human intelligence—sometimes called "artificial general intelligence" (AGI)—is for mathematical reasons impossible. The argument for this thesis rests on the fact that (1.) human intelligence is a capability of a complex dynamic system—the human brain and central nervous system, and (2.) systems of this sort (like all organic systems) cannot be modelled mathematically in a way that would allow the models to operate inside a computer. We survey on this basis the potential of AI in the future of clinical and translational research. <br />
<br />
'''--break--'''<br />
<br />
'''11:00-12:30p''' Working sessions/Presentations<br />
<br />
*'''Discussion and brief presentations. Topics to include:<br />
:The Clinical and Translational Science ontology landscape in 2023<br />
:Opportunities for AI in medicine<br />
:Will AI make ontologies redundant?<br />
:TBD<br />
<br />
'''12:30-01:30p''' Networking and Working Lunch <br />
<br />
'''01:30-3:00p''' Working sessions/Presentations<br />
<br />
*'''Yonghui Wu:''' A large language model for electronic health records.<br />
<br />
Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
'''--break--'''<br />
<br />
'''3:15-5:00p''' Working sessions/Presentations<br />
<br />
*'''Richard Ohrbach and Barry Smith:''' Defining 'Injury'<br />
<br />
This session is intended as a contribution to the ontology component of a broader investigation of injury and pain. SNOMED-CT defines an injury as a 'disorder resulting from physical damage to the body'. The WHO defines an injury as a 'bodily lesion at the organic level, resulting from acute exposure to energy (mechanical, thermal, electrical, chemical or radiant), in amounts that exceed the threshold of physiological tolerance.' We will explore these and other definitions with a view to establishing a more coherent understanding of the ontology of injury and of related phenomena such as lesion, trauma, pain, and so forth. Further topics for possible discussion are listed <u>[[Injury | here]]</u>. For the ontology of pain, see in particular [https://philpapers.org/archive/SMITAO-12.pdf here]<br />
<br />
'''--break--'''<br />
<br />
'''5:30p''' Networking Reception<br />
<br />
=='''Agenda: Friday Feb 24'''==<br />
<br />
'''08:30-09:15a''' Breakfast and Networking<br />
<br />
'''09:15-10:15a''' Working sessions/Presentations <br />
<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health.<br />
<br />
The ontological representation of occupation presents several challenges. For example, is it your occupation if you have not yet done it? Are you a nuclear engineer before you start your first job doing it? Also, if an occupation is a role (in the BFO sense), then how does it differ from a job role? If I have an occupation of professor, and have had four job roles as professors at different organizations, then that is one occupation numerically but four job roles. How to handle non-employment situations such as "retired" and "homemaker"? What if you concurrently hold two jobs in different "occupations"? I will discuss these issues in the context of recent discussions at the OMRSE monthly meeting and present some options for ontological representation of occupation and related entities.<br />
<br />
'''--break--'''<br />
<br />
'''10:30-12:30p''' Working sessions/Presentations <br />
<br />
*'''Justin Reese:''' Unsupervised machine learning to define subtypes of long COVID using the Human Phenotype Ontology. [[Authors | Co-authors]] <br />
<br />
Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterized by a wide range of manifestations that are difficult to analyze computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.<br />
<br />
We present a method for defining subtypes of long COVID by computationally modeling PASC phenotype data and applying unsupervised machine learning. We extracted Long COVID phenotype data expressed as HPO (Human Phenotype Ontology) terms from electronic healthcare records (EHRs), and then used semantic similarity of phenotype data to calculate a matrix of pairwise patient similarity. This matrix was then used to clustered patients using unsupervised machine learning (k means clustering).<br />
<br />
We identified six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centers to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems.<br />
<br />
Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.<br />
<br />
*'''Paul Heider:''' Beyond Structure: Using Standardized Labels from a Clinical Section Ontology in Natural Language Processing.<br />
<br />
The structured subdivision of clinical notes into sections is often treated as the primary and ultimate use for automatically tagging section boundaries. We discuss how section information (especially when normalized to a standard ontology) can be used to augment natural language processing algorithms, simplify information retrieval tasks,and speed up system development through disaggregated performance evaluation. Our examples emphasize a Clinical Section Ontology recently developed at MUSC but are applicable to any standardized section ontology. We also include a brief overview of the annotated datasets in this domain.<br />
<br />
'''12:30-01:30p''' Networking and Working Lunch <br />
<br />
'''1:00-03:15p''' Working sessions/Presentations from industry partners<br />
<br />
*'''Andy Lin: (VP Strategy, CTO | Mark III Systems)''' NVIDIA DGX H100 – The Premier AI Platform for Research and LLMs.<br />
<br />
With up to 4x training and up to 30x inferencing performance improvements for Large Language Models (LLMs) vs the already industry leading A100 GPU, the recently launched H100 GPU (Hopper) and its purpose-built DGX H100 AI platform are transforming what’s possible with Research and LLMs. In this session, we’ll dive into the DGX H100 platform and how you can get started or accelerate your work with NVIDIA and Mark III.<br />
<br />
*'''John Doole: (TriNetX, LLC)''' Medication Informatics at TriNetX: Powering Clinical Research with a Federated Real World Data Network.<br />
<br />
In this talk we will highlight the importance of medication informatics in clinical research, particularly in the context of a federated network. The talk discusses the challenges of data standardization and interoperability and the benefits of using medication ontologies and classification systems to improve data quality and efficiency. Additionally, the talk presents TriNetX's medication framework and how natural language processing is used to extract medication details, showcasing examples of its implementation and a discussion of the advantages and disadvantages of using NLP in clinical research.<br />
<br />
'''03:15-3:30p''' Closing Remarks<br />
<br />
== '''Participants''' ==<br />
<br />
* Alekseyenko, Alex, Medical University of South Carolina<br />
* Baker, Hamilton, Medical University of South Carolina<br />
* Doole, John, TriNetX<br />
* Heider, Paul, Medical University of South Carolina<br />
* Landgrebe, Jobst, Cognotekt, Cologne<br />
* Hogan, William, co-organizer, University of Florida<br />
* Hutchinson, Tom, University of Pennsylvania, Institute for BioInformatics<br />
* Masci, Anna Maria, National Institute of Environmental Health Sciences (NIEHS)<br />
* Obeid, Jihad, co-organizer, Medical University of South Carolina<br />
* Ohrbach, Richard, University at Buffalo<br />
* Reese, Justin, Lawrence Berkeley National Laboratory<br />
* Scheuermann, Richard, J. Craig Venter Institute<br />
* Simpson, Kit, Medical University of South Carolina<br />
* Skowronek, Matt, TriNetX<br />
* Smith, Barry, co-organizer, University at Buffalo<br />
* Topaloglu, Umit, Center for Biomedical Informatics and Information Technology (CBIIT) National Cancer Institute<br />
* Wehbe, Ramsey, Medical University of South Carolina<br />
* Yonghui Wu, University of Florida<br />
* Chatterjee, Prosenjit, The Citadel<br />
* Zheng, Jie, University of Pennsylvania</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73420CTS Ontology Workshop 20232023-02-22T16:21:49Z<p>Jihad: /* Agenda: Thursday Feb 23 */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
[[File:Chs23sm.jpg|link=|right|alt=Charleston]]<br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
* Social Determinants of Health<br />
* Mental health<br />
* ChatGPT<br />
<br />
== '''Organizers''' ==<br />
<br />
'''Workshop Co-organizers'''<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
'''CTSOG Co-chairs'''<br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* '''Medical University of South Carolina, the Biomedical Informatics Center''' and:<br />
<br />
[[File:Markiii-logo_160.png|link=https://www.markiiisys.com/|Mark III Systems]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:Nvidia-logo_180.png|link=https://www.nvidia.com/en-us/|NVIDIA]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:TriNetX-Logo-Horizontal-White_180.jpg|link=http://trinetx.com/|TriNetX]]<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Agenda: Thursday Feb 23'''==<br />
<br />
'''08:30a''': Breakfast and Networking<br />
<br />
'''09:00a''': Introduction to the meeting<br />
<br />
'''09:15-10:45a''' Working sessions/Presentations<br />
<br />
*'''Barry Smith and Jobst Landgrebe:''' Discussion on AI and Medicine, with a special focus on ChatGPT<br />
<br />
How are we to explain the peculiar tendency of ChatGPT to throw up what are called [https://buffalo.box.com/v/ChatGPT-Hallucinations ‘hallucinations’]? To answer this question we will draw on our book ''[https://buffalo.app.box.com/v/AI-Without-Fear Why Machines Will Never Rule the World]'', whose core thesis is that an artificial intelligence that could equal or exceed human intelligence—sometimes called "artificial general intelligence" (AGI)—is for mathematical reasons impossible. The argument for this thesis rests on the fact that (1.) human intelligence is a capability of a complex dynamic system—the human brain and central nervous system, and (2.) systems of this sort (like all organic systems) cannot be modelled mathematically in a way that would allow the models to operate inside a computer. We survey on this basis the potential of AI in the future of clinical and translational research. <br />
<br />
'''--break--'''<br />
<br />
'''11:00-12:30p''' Working sessions/Presentations<br />
<br />
*'''Discussion and brief presentations. Topics to include:<br />
:The Clinical and Translational Science ontology landscape in 2023<br />
:Opportunities for AI in medicine<br />
:Will AI make ontologies redundant?<br />
:TBD<br />
<br />
'''12:30-01:30p''' Networking and Working Lunch <br />
<br />
'''01:30-3:00p''' Working sessions/Presentations<br />
<br />
*'''Yonghui Wu:''' A large language model for electronic health records.<br />
<br />
Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
'''--break--'''<br />
<br />
'''3:15-5:00p''' Working sessions/Presentations<br />
<br />
*'''Richard Ohrbach and Barry Smith:''' Defining 'Injury'<br />
<br />
This session is intended as a contribution to the ontology component of a broader investigation of injury and pain. SNOMED-CT defines an injury as a 'disorder resulting from physical damage to the body'. The WHO defines an injury as a 'bodily lesion at the organic level, resulting from acute exposure to energy (mechanical, thermal, electrical, chemical or radiant), in amounts that exceed the threshold of physiological tolerance.' We will explore these and other definitions with a view to establishing a more coherent understanding of the ontology of injury and of related phenomena such as lesion, trauma, pain, and so forth. Further topics for possible discussion are listed <u>[[Injury | here]]</u>. For the ontology of pain, see in particular [https://philpapers.org/archive/SMITAO-12.pdf here]<br />
<br />
'''--break--'''<br />
<br />
'''5:30p''' Networking Reception<br />
<br />
=='''Agenda: Friday Feb 24'''==<br />
<br />
'''08:30-09:15a''' Breakfast and Networking<br />
<br />
'''09:15-10:15a''' Working sessions/Presentations <br />
<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health.<br />
<br />
The ontological representation of occupation presents several challenges. For example, is it your occupation if you have not yet done it? Are you a nuclear engineer before you start your first job doing it? Also, if an occupation is a role (in the BFO sense), then how does it differ from a job role? If I have an occupation of professor, and have had four job roles as professors at different organizations, then that is one occupation numerically but four job roles. How to handle non-employment situations such as "retired" and "homemaker"? What if you concurrently hold two jobs in different "occupations"? I will discuss these issues in the context of recent discussions at the OMRSE monthly meeting and present some options for ontological representation of occupation and related entities.<br />
<br />
'''--break--'''<br />
<br />
'''10:30-12:30p''' Working sessions/Presentations <br />
<br />
*'''Justin Reese:''' Unsupervised machine learning to define subtypes of long COVID using the Human Phenotype Ontology. [[Authors | Co-authors]] <br />
<br />
Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterized by a wide range of manifestations that are difficult to analyze computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.<br />
<br />
We present a method for defining subtypes of long COVID by computationally modeling PASC phenotype data and applying unsupervised machine learning. We extracted Long COVID phenotype data expressed as HPO (Human Phenotype Ontology) terms from electronic healthcare records (EHRs), and then used semantic similarity of phenotype data to calculate a matrix of pairwise patient similarity. This matrix was then used to clustered patients using unsupervised machine learning (k means clustering).<br />
<br />
We identified six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centers to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems.<br />
<br />
Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.<br />
<br />
*'''Paul Heider:''' Beyond Structure: Using Standardized Labels from a Clinical Section Ontology in Natural Language Processing.<br />
<br />
The structured subdivision of clinical notes into sections is often treated as the primary and ultimate use for automatically tagging section boundaries. We discuss how section information (especially when normalized to a standard ontology) can be used to augment natural language processing algorithms, simplify information retrieval tasks,and speed up system development through disaggregated performance evaluation. Our examples emphasize a Clinical Section Ontology recently developed at MUSC but are applicable to any standardized section ontology. We also include a brief overview of the annotated datasets in this domain.<br />
<br />
'''12:30-01:30p''' Networking and Working Lunch <br />
<br />
'''1:00-03:15p''' Working sessions/Presentations from industry partners<br />
<br />
*'''Andy Lin: (VP Strategy, CTO | Mark III Systems)''' NVIDIA DGX H100 – The Premier AI Platform for Research and LLMs.<br />
<br />
With up to 4x training and up to 30x inferencing performance improvements for Large Language Models (LLMs) vs the already industry leading A100 GPU, the recently launched H100 GPU (Hopper) and its purpose-built DGX H100 AI platform are transforming what’s possible with Research and LLMs. In this session, we’ll dive into the DGX H100 platform and how you can get started or accelerate your work with NVIDIA and Mark III.<br />
<br />
*'''John Doole: (TriNetX, LLC)''' Medication Informatics at TriNetX: Powering Clinical Research with a Federated Real World Data Network.<br />
<br />
In this talk we will highlight the importance of medication informatics in clinical research, particularly in the context of a federated network. The talk discusses the challenges of data standardization and interoperability and the benefits of using medication ontologies and classification systems to improve data quality and efficiency. Additionally, the talk presents TriNetX's medication framework and how natural language processing is used to extract medication details, showcasing examples of its implementation and a discussion of the advantages and disadvantages of using NLP in clinical research.<br />
<br />
'''03:15-3:30p''' Closing Remarks<br />
<br />
== '''Participants''' ==<br />
<br />
* Alekseyenko, Alex, Medical University of South Carolina<br />
* Baker, Hamilton, Medical University of South Carolina<br />
* Doole, John, TriNetX<br />
* Heider, Paul, Medical University of South Carolina<br />
* Landgrebe, Jobst, Cognotekt, Cologne<br />
* Hogan, William, co-organizer, University of Florida<br />
* Hutchinson, Tom, University of Pennsylvania, Institute for BioInformatics<br />
* Masci, Anna Maria, National Institute of Environmental Health Sciences (NIEHS)<br />
* Obeid, Jihad, co-organizer, Medical University of South Carolina<br />
* Ohrbach, Richard, University at Buffalo<br />
* Reese, Justin, Lawrence Berkeley National Laboratory<br />
* Scheuermann, Richard, J. Craig Venter Institute<br />
* Simpson, Kit, Medical University of South Carolina<br />
* Skowronek, Matt, TriNetX<br />
* Smith, Barry, co-organizer, University at Buffalo<br />
* Topaloglu, Umit, Center for Biomedical Informatics and Information Technology (CBIIT) National Cancer Institute<br />
* Wehbe, Ramsey, Medical University of South Carolina<br />
* Yonghui Wu, University of Florida</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73419CTS Ontology Workshop 20232023-02-22T16:20:21Z<p>Jihad: /* Draft Agenda: Friday Feb 24 */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
[[File:Chs23sm.jpg|link=|right|alt=Charleston]]<br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
* Social Determinants of Health<br />
* Mental health<br />
* ChatGPT<br />
<br />
== '''Organizers''' ==<br />
<br />
'''Workshop Co-organizers'''<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
'''CTSOG Co-chairs'''<br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* '''Medical University of South Carolina, the Biomedical Informatics Center''' and:<br />
<br />
[[File:Markiii-logo_160.png|link=https://www.markiiisys.com/|Mark III Systems]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:Nvidia-logo_180.png|link=https://www.nvidia.com/en-us/|NVIDIA]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:TriNetX-Logo-Horizontal-White_180.jpg|link=http://trinetx.com/|TriNetX]]<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Agenda: Thursday Feb 23'''==<br />
<br />
'''08:30a''': Breakfast and Networking<br />
<br />
'''09:00a''': Introduction to the meeting<br />
<br />
'''09:15-10:45a''' Working sessions/Presentations<br />
<br />
*'''Barry Smith and Jobst Landgrebe:''' Discussion on AI and Medicine, with a special focus on ChatGPT<br />
<br />
How are we to explain the peculiar tendency of ChatGPT to throw up what are called [https://buffalo.box.com/v/ChatGPT-Hallucinations ‘hallucinations’]? To answer this question we will draw on our book ''[https://buffalo.app.box.com/v/AI-Without-Fear Why Machines Will Never Rule the World]'', whose core thesis is that an artificial intelligence that could equal or exceed human intelligence—sometimes called "artificial general intelligence" (AGI)—is for mathematical reasons impossible. The argument for this thesis rests on the fact that (1.) human intelligence is a capability of a complex dynamic system—the human brain and central nervous system, and (2.) systems of this sort (like all organic systems) cannot be modelled mathematically in a way that would allow the models to operate inside a computer. We survey on this basis the potential of AI in the future of clinical and translational research. <br />
<br />
'''--break--'''<br />
<br />
'''11:00-12:30p''' Working sessions/Presentations<br />
<br />
*'''Discussion and brief presentations. Topics to include:<br />
:The Clinical and Translational Science ontology landscape in 2023<br />
:Opportunities for AI in medicine<br />
:Will AI make ontologies redundant?<br />
:TBD<br />
<br />
'''12:30-01:30p''' Networking and Working Lunch <br />
<br />
'''01:30-3:00p''' Working sessions/Presentations<br />
<br />
*'''Yonghui Wu:''' A large language model for electronic health records.<br />
<br />
Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
'''--break--'''<br />
<br />
'''3:15-5:00p''' Working sessions/Presentations<br />
<br />
*'''Richard Ohrbach and Barry Smith:''' Defining 'Injury'<br />
<br />
This session is intended as a contribution to the ontology component of a broader investigation of injury and pain. SNOMED-CT defines an injury as a 'disorder resulting from physical damage to the body'. The WHO defines an injury as a 'bodily lesion at the organic level, resulting from acute exposure to energy (mechanical, thermal, electrical, chemical or radiant), in amounts that exceed the threshold of physiological tolerance.' We will explore these and other definitions with a view to establishing a more coherent understanding of the ontology of injury and of related phenomena such as lesion, trauma, pain, and so forth. Further topics for possible discussion are listed <u>[[Injury | here]]</u>. For the ontology of pain, see in particular [https://philpapers.org/archive/SMITAO-12.pdf here]<br />
<br />
=='''Agenda: Friday Feb 24'''==<br />
<br />
'''08:30-09:15a''' Breakfast and Networking<br />
<br />
'''09:15-10:15a''' Working sessions/Presentations <br />
<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health.<br />
<br />
The ontological representation of occupation presents several challenges. For example, is it your occupation if you have not yet done it? Are you a nuclear engineer before you start your first job doing it? Also, if an occupation is a role (in the BFO sense), then how does it differ from a job role? If I have an occupation of professor, and have had four job roles as professors at different organizations, then that is one occupation numerically but four job roles. How to handle non-employment situations such as "retired" and "homemaker"? What if you concurrently hold two jobs in different "occupations"? I will discuss these issues in the context of recent discussions at the OMRSE monthly meeting and present some options for ontological representation of occupation and related entities.<br />
<br />
'''--break--'''<br />
<br />
'''10:30-12:30p''' Working sessions/Presentations <br />
<br />
*'''Justin Reese:''' Unsupervised machine learning to define subtypes of long COVID using the Human Phenotype Ontology. [[Authors | Co-authors]] <br />
<br />
Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterized by a wide range of manifestations that are difficult to analyze computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.<br />
<br />
We present a method for defining subtypes of long COVID by computationally modeling PASC phenotype data and applying unsupervised machine learning. We extracted Long COVID phenotype data expressed as HPO (Human Phenotype Ontology) terms from electronic healthcare records (EHRs), and then used semantic similarity of phenotype data to calculate a matrix of pairwise patient similarity. This matrix was then used to clustered patients using unsupervised machine learning (k means clustering).<br />
<br />
We identified six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centers to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems.<br />
<br />
Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.<br />
<br />
*'''Paul Heider:''' Beyond Structure: Using Standardized Labels from a Clinical Section Ontology in Natural Language Processing.<br />
<br />
The structured subdivision of clinical notes into sections is often treated as the primary and ultimate use for automatically tagging section boundaries. We discuss how section information (especially when normalized to a standard ontology) can be used to augment natural language processing algorithms, simplify information retrieval tasks,and speed up system development through disaggregated performance evaluation. Our examples emphasize a Clinical Section Ontology recently developed at MUSC but are applicable to any standardized section ontology. We also include a brief overview of the annotated datasets in this domain.<br />
<br />
'''12:30-01:30p''' Networking and Working Lunch <br />
<br />
'''1:00-03:15p''' Working sessions/Presentations from industry partners<br />
<br />
*'''Andy Lin: (VP Strategy, CTO | Mark III Systems)''' NVIDIA DGX H100 – The Premier AI Platform for Research and LLMs.<br />
<br />
With up to 4x training and up to 30x inferencing performance improvements for Large Language Models (LLMs) vs the already industry leading A100 GPU, the recently launched H100 GPU (Hopper) and its purpose-built DGX H100 AI platform are transforming what’s possible with Research and LLMs. In this session, we’ll dive into the DGX H100 platform and how you can get started or accelerate your work with NVIDIA and Mark III.<br />
<br />
*'''John Doole: (TriNetX, LLC)''' Medication Informatics at TriNetX: Powering Clinical Research with a Federated Real World Data Network.<br />
<br />
In this talk we will highlight the importance of medication informatics in clinical research, particularly in the context of a federated network. The talk discusses the challenges of data standardization and interoperability and the benefits of using medication ontologies and classification systems to improve data quality and efficiency. Additionally, the talk presents TriNetX's medication framework and how natural language processing is used to extract medication details, showcasing examples of its implementation and a discussion of the advantages and disadvantages of using NLP in clinical research.<br />
<br />
'''03:15-3:30p''' Closing Remarks<br />
<br />
== '''Participants''' ==<br />
<br />
* Alekseyenko, Alex, Medical University of South Carolina<br />
* Baker, Hamilton, Medical University of South Carolina<br />
* Doole, John, TriNetX<br />
* Heider, Paul, Medical University of South Carolina<br />
* Landgrebe, Jobst, Cognotekt, Cologne<br />
* Hogan, William, co-organizer, University of Florida<br />
* Hutchinson, Tom, University of Pennsylvania, Institute for BioInformatics<br />
* Masci, Anna Maria, National Institute of Environmental Health Sciences (NIEHS)<br />
* Obeid, Jihad, co-organizer, Medical University of South Carolina<br />
* Ohrbach, Richard, University at Buffalo<br />
* Reese, Justin, Lawrence Berkeley National Laboratory<br />
* Scheuermann, Richard, J. Craig Venter Institute<br />
* Simpson, Kit, Medical University of South Carolina<br />
* Skowronek, Matt, TriNetX<br />
* Smith, Barry, co-organizer, University at Buffalo<br />
* Topaloglu, Umit, Center for Biomedical Informatics and Information Technology (CBIIT) National Cancer Institute<br />
* Wehbe, Ramsey, Medical University of South Carolina<br />
* Yonghui Wu, University of Florida</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73418CTS Ontology Workshop 20232023-02-22T16:20:11Z<p>Jihad: /* Draft Agenda: Thursday Feb 23 */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
[[File:Chs23sm.jpg|link=|right|alt=Charleston]]<br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
* Social Determinants of Health<br />
* Mental health<br />
* ChatGPT<br />
<br />
== '''Organizers''' ==<br />
<br />
'''Workshop Co-organizers'''<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
'''CTSOG Co-chairs'''<br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* '''Medical University of South Carolina, the Biomedical Informatics Center''' and:<br />
<br />
[[File:Markiii-logo_160.png|link=https://www.markiiisys.com/|Mark III Systems]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:Nvidia-logo_180.png|link=https://www.nvidia.com/en-us/|NVIDIA]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:TriNetX-Logo-Horizontal-White_180.jpg|link=http://trinetx.com/|TriNetX]]<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Agenda: Thursday Feb 23'''==<br />
<br />
'''08:30a''': Breakfast and Networking<br />
<br />
'''09:00a''': Introduction to the meeting<br />
<br />
'''09:15-10:45a''' Working sessions/Presentations<br />
<br />
*'''Barry Smith and Jobst Landgrebe:''' Discussion on AI and Medicine, with a special focus on ChatGPT<br />
<br />
How are we to explain the peculiar tendency of ChatGPT to throw up what are called [https://buffalo.box.com/v/ChatGPT-Hallucinations ‘hallucinations’]? To answer this question we will draw on our book ''[https://buffalo.app.box.com/v/AI-Without-Fear Why Machines Will Never Rule the World]'', whose core thesis is that an artificial intelligence that could equal or exceed human intelligence—sometimes called "artificial general intelligence" (AGI)—is for mathematical reasons impossible. The argument for this thesis rests on the fact that (1.) human intelligence is a capability of a complex dynamic system—the human brain and central nervous system, and (2.) systems of this sort (like all organic systems) cannot be modelled mathematically in a way that would allow the models to operate inside a computer. We survey on this basis the potential of AI in the future of clinical and translational research. <br />
<br />
'''--break--'''<br />
<br />
'''11:00-12:30p''' Working sessions/Presentations<br />
<br />
*'''Discussion and brief presentations. Topics to include:<br />
:The Clinical and Translational Science ontology landscape in 2023<br />
:Opportunities for AI in medicine<br />
:Will AI make ontologies redundant?<br />
:TBD<br />
<br />
'''12:30-01:30p''' Networking and Working Lunch <br />
<br />
'''01:30-3:00p''' Working sessions/Presentations<br />
<br />
*'''Yonghui Wu:''' A large language model for electronic health records.<br />
<br />
Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
'''--break--'''<br />
<br />
'''3:15-5:00p''' Working sessions/Presentations<br />
<br />
*'''Richard Ohrbach and Barry Smith:''' Defining 'Injury'<br />
<br />
This session is intended as a contribution to the ontology component of a broader investigation of injury and pain. SNOMED-CT defines an injury as a 'disorder resulting from physical damage to the body'. The WHO defines an injury as a 'bodily lesion at the organic level, resulting from acute exposure to energy (mechanical, thermal, electrical, chemical or radiant), in amounts that exceed the threshold of physiological tolerance.' We will explore these and other definitions with a view to establishing a more coherent understanding of the ontology of injury and of related phenomena such as lesion, trauma, pain, and so forth. Further topics for possible discussion are listed <u>[[Injury | here]]</u>. For the ontology of pain, see in particular [https://philpapers.org/archive/SMITAO-12.pdf here]<br />
<br />
=='''Draft Agenda: Friday Feb 24'''==<br />
<br />
'''08:30-09:15a''' Breakfast and Networking<br />
<br />
'''09:15-10:15a''' Working sessions/Presentations <br />
<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health.<br />
<br />
The ontological representation of occupation presents several challenges. For example, is it your occupation if you have not yet done it? Are you a nuclear engineer before you start your first job doing it? Also, if an occupation is a role (in the BFO sense), then how does it differ from a job role? If I have an occupation of professor, and have had four job roles as professors at different organizations, then that is one occupation numerically but four job roles. How to handle non-employment situations such as "retired" and "homemaker"? What if you concurrently hold two jobs in different "occupations"? I will discuss these issues in the context of recent discussions at the OMRSE monthly meeting and present some options for ontological representation of occupation and related entities.<br />
<br />
'''--break--'''<br />
<br />
'''10:30-12:30p''' Working sessions/Presentations <br />
<br />
*'''Justin Reese:''' Unsupervised machine learning to define subtypes of long COVID using the Human Phenotype Ontology. [[Authors | Co-authors]] <br />
<br />
Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterized by a wide range of manifestations that are difficult to analyze computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.<br />
<br />
We present a method for defining subtypes of long COVID by computationally modeling PASC phenotype data and applying unsupervised machine learning. We extracted Long COVID phenotype data expressed as HPO (Human Phenotype Ontology) terms from electronic healthcare records (EHRs), and then used semantic similarity of phenotype data to calculate a matrix of pairwise patient similarity. This matrix was then used to clustered patients using unsupervised machine learning (k means clustering).<br />
<br />
We identified six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centers to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems.<br />
<br />
Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.<br />
<br />
*'''Paul Heider:''' Beyond Structure: Using Standardized Labels from a Clinical Section Ontology in Natural Language Processing.<br />
<br />
The structured subdivision of clinical notes into sections is often treated as the primary and ultimate use for automatically tagging section boundaries. We discuss how section information (especially when normalized to a standard ontology) can be used to augment natural language processing algorithms, simplify information retrieval tasks,and speed up system development through disaggregated performance evaluation. Our examples emphasize a Clinical Section Ontology recently developed at MUSC but are applicable to any standardized section ontology. We also include a brief overview of the annotated datasets in this domain.<br />
<br />
'''12:30-01:30p''' Networking and Working Lunch <br />
<br />
'''1:00-03:15p''' Working sessions/Presentations from industry partners<br />
<br />
*'''Andy Lin: (VP Strategy, CTO | Mark III Systems)''' NVIDIA DGX H100 – The Premier AI Platform for Research and LLMs.<br />
<br />
With up to 4x training and up to 30x inferencing performance improvements for Large Language Models (LLMs) vs the already industry leading A100 GPU, the recently launched H100 GPU (Hopper) and its purpose-built DGX H100 AI platform are transforming what’s possible with Research and LLMs. In this session, we’ll dive into the DGX H100 platform and how you can get started or accelerate your work with NVIDIA and Mark III.<br />
<br />
*'''John Doole: (TriNetX, LLC)''' Medication Informatics at TriNetX: Powering Clinical Research with a Federated Real World Data Network.<br />
<br />
In this talk we will highlight the importance of medication informatics in clinical research, particularly in the context of a federated network. The talk discusses the challenges of data standardization and interoperability and the benefits of using medication ontologies and classification systems to improve data quality and efficiency. Additionally, the talk presents TriNetX's medication framework and how natural language processing is used to extract medication details, showcasing examples of its implementation and a discussion of the advantages and disadvantages of using NLP in clinical research.<br />
<br />
'''03:15-3:30p''' Closing Remarks<br />
<br />
== '''Participants''' ==<br />
<br />
* Alekseyenko, Alex, Medical University of South Carolina<br />
* Baker, Hamilton, Medical University of South Carolina<br />
* Doole, John, TriNetX<br />
* Heider, Paul, Medical University of South Carolina<br />
* Landgrebe, Jobst, Cognotekt, Cologne<br />
* Hogan, William, co-organizer, University of Florida<br />
* Hutchinson, Tom, University of Pennsylvania, Institute for BioInformatics<br />
* Masci, Anna Maria, National Institute of Environmental Health Sciences (NIEHS)<br />
* Obeid, Jihad, co-organizer, Medical University of South Carolina<br />
* Ohrbach, Richard, University at Buffalo<br />
* Reese, Justin, Lawrence Berkeley National Laboratory<br />
* Scheuermann, Richard, J. Craig Venter Institute<br />
* Simpson, Kit, Medical University of South Carolina<br />
* Skowronek, Matt, TriNetX<br />
* Smith, Barry, co-organizer, University at Buffalo<br />
* Topaloglu, Umit, Center for Biomedical Informatics and Information Technology (CBIIT) National Cancer Institute<br />
* Wehbe, Ramsey, Medical University of South Carolina<br />
* Yonghui Wu, University of Florida</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73417CTS Ontology Workshop 20232023-02-22T14:57:56Z<p>Jihad: /* Participants */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
[[File:Chs23sm.jpg|link=|right|alt=Charleston]]<br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
* Social Determinants of Health<br />
* Mental health<br />
* ChatGPT<br />
<br />
== '''Organizers''' ==<br />
<br />
'''Workshop Co-organizers'''<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
'''CTSOG Co-chairs'''<br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* '''Medical University of South Carolina, the Biomedical Informatics Center''' and:<br />
<br />
[[File:Markiii-logo_160.png|link=https://www.markiiisys.com/|Mark III Systems]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:Nvidia-logo_180.png|link=https://www.nvidia.com/en-us/|NVIDIA]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:TriNetX-Logo-Horizontal-White_180.jpg|link=http://trinetx.com/|TriNetX]]<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Draft Agenda: Thursday Feb 23'''==<br />
<br />
'''08:30a''': Breakfast and Networking<br />
<br />
'''09:00a''': Introduction to the meeting<br />
<br />
'''09:15-10:45a''' Working sessions/Presentations<br />
<br />
*'''Barry Smith and Jobst Landgrebe:''' Discussion on AI and Medicine, with a special focus on ChatGPT<br />
<br />
How are we to explain the peculiar tendency of ChatGPT to throw up what are called [https://buffalo.box.com/v/ChatGPT-Hallucinations ‘hallucinations’]? To answer this question we will draw on our book ''[https://buffalo.app.box.com/v/AI-Without-Fear Why Machines Will Never Rule the World]'', whose core thesis is that an artificial intelligence that could equal or exceed human intelligence—sometimes called "artificial general intelligence" (AGI)—is for mathematical reasons impossible. The argument for this thesis rests on the fact that (1.) human intelligence is a capability of a complex dynamic system—the human brain and central nervous system, and (2.) systems of this sort (like all organic systems) cannot be modelled mathematically in a way that would allow the models to operate inside a computer. We survey on this basis the potential of AI in the future of clinical and translational research. <br />
<br />
'''--break--'''<br />
<br />
'''11:00-12:30p''' Working sessions/Presentations<br />
<br />
*'''Discussion and brief presentations. Topics to include:<br />
:The Clinical and Translational Science ontology landscape in 2023<br />
:Opportunities for AI in medicine<br />
:Will AI make ontologies redundant?<br />
:TBD<br />
<br />
'''12:30-01:30p''' Networking and Working Lunch <br />
<br />
'''01:30-3:00p''' Working sessions/Presentations<br />
<br />
*'''Yonghui Wu:''' A large language model for electronic health records.<br />
<br />
Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
'''--break--'''<br />
<br />
'''3:15-5:00p''' Working sessions/Presentations<br />
<br />
*'''Richard Ohrbach and Barry Smith:''' Defining 'Injury'<br />
<br />
This session is intended as a contribution to the ontology component of a broader investigation of injury and pain. SNOMED-CT defines an injury as a 'disorder resulting from physical damage to the body'. The WHO defines an injury as a 'bodily lesion at the organic level, resulting from acute exposure to energy (mechanical, thermal, electrical, chemical or radiant), in amounts that exceed the threshold of physiological tolerance.' We will explore these and other definitions with a view to establishing a more coherent understanding of the ontology of injury and of related phenomena such as lesion, trauma, pain, and so forth. Further topics for possible discussion are listed <u>[[Injury | here]]</u>. For the ontology of pain, see in particular [https://philpapers.org/archive/SMITAO-12.pdf here]<br />
<br />
=='''Draft Agenda: Friday Feb 24'''==<br />
<br />
'''08:30-09:15a''' Breakfast and Networking<br />
<br />
'''09:15-10:15a''' Working sessions/Presentations <br />
<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health.<br />
<br />
The ontological representation of occupation presents several challenges. For example, is it your occupation if you have not yet done it? Are you a nuclear engineer before you start your first job doing it? Also, if an occupation is a role (in the BFO sense), then how does it differ from a job role? If I have an occupation of professor, and have had four job roles as professors at different organizations, then that is one occupation numerically but four job roles. How to handle non-employment situations such as "retired" and "homemaker"? What if you concurrently hold two jobs in different "occupations"? I will discuss these issues in the context of recent discussions at the OMRSE monthly meeting and present some options for ontological representation of occupation and related entities.<br />
<br />
'''--break--'''<br />
<br />
'''10:30-12:30p''' Working sessions/Presentations <br />
<br />
*'''Justin Reese:''' Unsupervised machine learning to define subtypes of long COVID using the Human Phenotype Ontology. [[Authors | Co-authors]] <br />
<br />
Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterized by a wide range of manifestations that are difficult to analyze computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.<br />
<br />
We present a method for defining subtypes of long COVID by computationally modeling PASC phenotype data and applying unsupervised machine learning. We extracted Long COVID phenotype data expressed as HPO (Human Phenotype Ontology) terms from electronic healthcare records (EHRs), and then used semantic similarity of phenotype data to calculate a matrix of pairwise patient similarity. This matrix was then used to clustered patients using unsupervised machine learning (k means clustering).<br />
<br />
We identified six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centers to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems.<br />
<br />
Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.<br />
<br />
*'''Paul Heider:''' Beyond Structure: Using Standardized Labels from a Clinical Section Ontology in Natural Language Processing.<br />
<br />
The structured subdivision of clinical notes into sections is often treated as the primary and ultimate use for automatically tagging section boundaries. We discuss how section information (especially when normalized to a standard ontology) can be used to augment natural language processing algorithms, simplify information retrieval tasks,and speed up system development through disaggregated performance evaluation. Our examples emphasize a Clinical Section Ontology recently developed at MUSC but are applicable to any standardized section ontology. We also include a brief overview of the annotated datasets in this domain.<br />
<br />
'''12:30-01:30p''' Networking and Working Lunch <br />
<br />
'''1:00-03:15p''' Working sessions/Presentations from industry partners<br />
<br />
*'''Andy Lin: (VP Strategy, CTO | Mark III Systems)''' NVIDIA DGX H100 – The Premier AI Platform for Research and LLMs.<br />
<br />
With up to 4x training and up to 30x inferencing performance improvements for Large Language Models (LLMs) vs the already industry leading A100 GPU, the recently launched H100 GPU (Hopper) and its purpose-built DGX H100 AI platform are transforming what’s possible with Research and LLMs. In this session, we’ll dive into the DGX H100 platform and how you can get started or accelerate your work with NVIDIA and Mark III.<br />
<br />
*'''John Doole: (TriNetX, LLC)''' Medication Informatics at TriNetX: Powering Clinical Research with a Federated Real World Data Network.<br />
<br />
In this talk we will highlight the importance of medication informatics in clinical research, particularly in the context of a federated network. The talk discusses the challenges of data standardization and interoperability and the benefits of using medication ontologies and classification systems to improve data quality and efficiency. Additionally, the talk presents TriNetX's medication framework and how natural language processing is used to extract medication details, showcasing examples of its implementation and a discussion of the advantages and disadvantages of using NLP in clinical research.<br />
<br />
'''03:15-3:30p''' Closing Remarks<br />
<br />
== '''Participants''' ==<br />
<br />
* Alekseyenko, Alex, Medical University of South Carolina<br />
* Baker, Hamilton, Medical University of South Carolina<br />
* Doole, John, TriNetX<br />
* Heider, Paul, Medical University of South Carolina<br />
* Landgrebe, Jobst, Cognotekt, Cologne<br />
* Hogan, William, co-organizer, University of Florida<br />
* Hutchinson, Tom, University of Pennsylvania, Institute for BioInformatics<br />
* Masci, Anna Maria, National Institute of Environmental Health Sciences (NIEHS)<br />
* Obeid, Jihad, co-organizer, Medical University of South Carolina<br />
* Ohrbach, Richard, University at Buffalo<br />
* Reese, Justin, Lawrence Berkeley National Laboratory<br />
* Scheuermann, Richard, J. Craig Venter Institute<br />
* Simpson, Kit, Medical University of South Carolina<br />
* Skowronek, Matt, TriNetX<br />
* Smith, Barry, co-organizer, University at Buffalo<br />
* Topaloglu, Umit, Center for Biomedical Informatics and Information Technology (CBIIT) National Cancer Institute<br />
* Wehbe, Ramsey, Medical University of South Carolina<br />
* Yonghui Wu, University of Florida</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73416CTS Ontology Workshop 20232023-02-22T14:44:29Z<p>Jihad: /* Draft Agenda: Friday Feb 24 */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
[[File:Chs23sm.jpg|link=|right|alt=Charleston]]<br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
* Social Determinants of Health<br />
* Mental health<br />
* ChatGPT<br />
<br />
== '''Organizers''' ==<br />
<br />
'''Workshop Co-organizers'''<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
'''CTSOG Co-chairs'''<br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* '''Medical University of South Carolina, the Biomedical Informatics Center''' and:<br />
<br />
[[File:Markiii-logo_160.png|link=https://www.markiiisys.com/|Mark III Systems]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:Nvidia-logo_180.png|link=https://www.nvidia.com/en-us/|NVIDIA]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:TriNetX-Logo-Horizontal-White_180.jpg|link=http://trinetx.com/|TriNetX]]<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Draft Agenda: Thursday Feb 23'''==<br />
<br />
'''08:30a''': Breakfast and Networking<br />
<br />
'''09:00a''': Introduction to the meeting<br />
<br />
'''09:15-10:45a''' Working sessions/Presentations<br />
<br />
*'''Barry Smith and Jobst Landgrebe:''' Discussion on AI and Medicine, with a special focus on ChatGPT<br />
<br />
How are we to explain the peculiar tendency of ChatGPT to throw up what are called [https://buffalo.box.com/v/ChatGPT-Hallucinations ‘hallucinations’]? To answer this question we will draw on our book ''[https://buffalo.app.box.com/v/AI-Without-Fear Why Machines Will Never Rule the World]'', whose core thesis is that an artificial intelligence that could equal or exceed human intelligence—sometimes called "artificial general intelligence" (AGI)—is for mathematical reasons impossible. The argument for this thesis rests on the fact that (1.) human intelligence is a capability of a complex dynamic system—the human brain and central nervous system, and (2.) systems of this sort (like all organic systems) cannot be modelled mathematically in a way that would allow the models to operate inside a computer. We survey on this basis the potential of AI in the future of clinical and translational research. <br />
<br />
'''--break--'''<br />
<br />
'''11:00-12:30p''' Working sessions/Presentations<br />
<br />
*'''Discussion and brief presentations. Topics to include:<br />
:The Clinical and Translational Science ontology landscape in 2023<br />
:Opportunities for AI in medicine<br />
:Will AI make ontologies redundant?<br />
:TBD<br />
<br />
'''12:30-01:30p''' Networking and Working Lunch <br />
<br />
'''01:30-3:00p''' Working sessions/Presentations<br />
<br />
*'''Yonghui Wu:''' A large language model for electronic health records.<br />
<br />
Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
'''--break--'''<br />
<br />
'''3:15-5:00p''' Working sessions/Presentations<br />
<br />
*'''Richard Ohrbach and Barry Smith:''' Defining 'Injury'<br />
<br />
This session is intended as a contribution to the ontology component of a broader investigation of injury and pain. SNOMED-CT defines an injury as a 'disorder resulting from physical damage to the body'. The WHO defines an injury as a 'bodily lesion at the organic level, resulting from acute exposure to energy (mechanical, thermal, electrical, chemical or radiant), in amounts that exceed the threshold of physiological tolerance.' We will explore these and other definitions with a view to establishing a more coherent understanding of the ontology of injury and of related phenomena such as lesion, trauma, pain, and so forth. Further topics for possible discussion are listed <u>[[Injury | here]]</u>. For the ontology of pain, see in particular [https://philpapers.org/archive/SMITAO-12.pdf here]<br />
<br />
=='''Draft Agenda: Friday Feb 24'''==<br />
<br />
'''08:30-09:15a''' Breakfast and Networking<br />
<br />
'''09:15-10:15a''' Working sessions/Presentations <br />
<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health.<br />
<br />
The ontological representation of occupation presents several challenges. For example, is it your occupation if you have not yet done it? Are you a nuclear engineer before you start your first job doing it? Also, if an occupation is a role (in the BFO sense), then how does it differ from a job role? If I have an occupation of professor, and have had four job roles as professors at different organizations, then that is one occupation numerically but four job roles. How to handle non-employment situations such as "retired" and "homemaker"? What if you concurrently hold two jobs in different "occupations"? I will discuss these issues in the context of recent discussions at the OMRSE monthly meeting and present some options for ontological representation of occupation and related entities.<br />
<br />
'''--break--'''<br />
<br />
'''10:30-12:30p''' Working sessions/Presentations <br />
<br />
*'''Justin Reese:''' Unsupervised machine learning to define subtypes of long COVID using the Human Phenotype Ontology. [[Authors | Co-authors]] <br />
<br />
Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterized by a wide range of manifestations that are difficult to analyze computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.<br />
<br />
We present a method for defining subtypes of long COVID by computationally modeling PASC phenotype data and applying unsupervised machine learning. We extracted Long COVID phenotype data expressed as HPO (Human Phenotype Ontology) terms from electronic healthcare records (EHRs), and then used semantic similarity of phenotype data to calculate a matrix of pairwise patient similarity. This matrix was then used to clustered patients using unsupervised machine learning (k means clustering).<br />
<br />
We identified six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centers to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems.<br />
<br />
Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.<br />
<br />
*'''Paul Heider:''' Beyond Structure: Using Standardized Labels from a Clinical Section Ontology in Natural Language Processing.<br />
<br />
The structured subdivision of clinical notes into sections is often treated as the primary and ultimate use for automatically tagging section boundaries. We discuss how section information (especially when normalized to a standard ontology) can be used to augment natural language processing algorithms, simplify information retrieval tasks,and speed up system development through disaggregated performance evaluation. Our examples emphasize a Clinical Section Ontology recently developed at MUSC but are applicable to any standardized section ontology. We also include a brief overview of the annotated datasets in this domain.<br />
<br />
'''12:30-01:30p''' Networking and Working Lunch <br />
<br />
'''1:00-03:15p''' Working sessions/Presentations from industry partners<br />
<br />
*'''Andy Lin: (VP Strategy, CTO | Mark III Systems)''' NVIDIA DGX H100 – The Premier AI Platform for Research and LLMs.<br />
<br />
With up to 4x training and up to 30x inferencing performance improvements for Large Language Models (LLMs) vs the already industry leading A100 GPU, the recently launched H100 GPU (Hopper) and its purpose-built DGX H100 AI platform are transforming what’s possible with Research and LLMs. In this session, we’ll dive into the DGX H100 platform and how you can get started or accelerate your work with NVIDIA and Mark III.<br />
<br />
*'''John Doole: (TriNetX, LLC)''' Medication Informatics at TriNetX: Powering Clinical Research with a Federated Real World Data Network.<br />
<br />
In this talk we will highlight the importance of medication informatics in clinical research, particularly in the context of a federated network. The talk discusses the challenges of data standardization and interoperability and the benefits of using medication ontologies and classification systems to improve data quality and efficiency. Additionally, the talk presents TriNetX's medication framework and how natural language processing is used to extract medication details, showcasing examples of its implementation and a discussion of the advantages and disadvantages of using NLP in clinical research.<br />
<br />
'''03:15-3:30p''' Closing Remarks<br />
<br />
== '''Participants''' ==<br />
<br />
* Alekseyenko, Alex, Medical University of South Carolina<br />
* Baker, Hamilton, Medical University of South Carolina<br />
* Doole, John, TriNetX<br />
* Heider, Paul, Medical University of South Carolina<br />
* Landgrebe, Jobst, Cognotekt, Cologne<br />
* Hogan, William, co-organizer, University of Florida<br />
* Hutchinson, Tom, University of Pennsylvania, Institute for BioInformatics<br />
* Masci, Anna Maria, National Institute of Environmental Health Sciences (NIEHS)<br />
* Obeid, Jihad, co-organizer, Medical University of South Carolina<br />
* Ohrbach, Richard, University at Buffalo<br />
* Reese, Justin, Lawrence Berkeley National Laboratory<br />
* Scheuermann, Richard, J. Craig Venter Institute<br />
* Simpson, Kit, Medical University of South Carolina<br />
* Skowronek, Matt, TriNetX<br />
* Smith, Barry, co-organizer, University at Buffalo<br />
* Topaloglu, Umit, Center for Biomedical Informatics and Information Technology (CBIIT) National Cancer Institute<br />
* Yonghui Wu, University of Florida</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73415CTS Ontology Workshop 20232023-02-22T14:43:26Z<p>Jihad: /* Draft Agenda: Friday Feb 24 */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
[[File:Chs23sm.jpg|link=|right|alt=Charleston]]<br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
* Social Determinants of Health<br />
* Mental health<br />
* ChatGPT<br />
<br />
== '''Organizers''' ==<br />
<br />
'''Workshop Co-organizers'''<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
'''CTSOG Co-chairs'''<br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* '''Medical University of South Carolina, the Biomedical Informatics Center''' and:<br />
<br />
[[File:Markiii-logo_160.png|link=https://www.markiiisys.com/|Mark III Systems]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:Nvidia-logo_180.png|link=https://www.nvidia.com/en-us/|NVIDIA]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:TriNetX-Logo-Horizontal-White_180.jpg|link=http://trinetx.com/|TriNetX]]<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Draft Agenda: Thursday Feb 23'''==<br />
<br />
'''08:30a''': Breakfast and Networking<br />
<br />
'''09:00a''': Introduction to the meeting<br />
<br />
'''09:15-10:45a''' Working sessions/Presentations<br />
<br />
*'''Barry Smith and Jobst Landgrebe:''' Discussion on AI and Medicine, with a special focus on ChatGPT<br />
<br />
How are we to explain the peculiar tendency of ChatGPT to throw up what are called [https://buffalo.box.com/v/ChatGPT-Hallucinations ‘hallucinations’]? To answer this question we will draw on our book ''[https://buffalo.app.box.com/v/AI-Without-Fear Why Machines Will Never Rule the World]'', whose core thesis is that an artificial intelligence that could equal or exceed human intelligence—sometimes called "artificial general intelligence" (AGI)—is for mathematical reasons impossible. The argument for this thesis rests on the fact that (1.) human intelligence is a capability of a complex dynamic system—the human brain and central nervous system, and (2.) systems of this sort (like all organic systems) cannot be modelled mathematically in a way that would allow the models to operate inside a computer. We survey on this basis the potential of AI in the future of clinical and translational research. <br />
<br />
'''--break--'''<br />
<br />
'''11:00-12:30p''' Working sessions/Presentations<br />
<br />
*'''Discussion and brief presentations. Topics to include:<br />
:The Clinical and Translational Science ontology landscape in 2023<br />
:Opportunities for AI in medicine<br />
:Will AI make ontologies redundant?<br />
:TBD<br />
<br />
'''12:30-01:30p''' Networking and Working Lunch <br />
<br />
'''01:30-3:00p''' Working sessions/Presentations<br />
<br />
*'''Yonghui Wu:''' A large language model for electronic health records.<br />
<br />
Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
'''--break--'''<br />
<br />
'''3:15-5:00p''' Working sessions/Presentations<br />
<br />
*'''Richard Ohrbach and Barry Smith:''' Defining 'Injury'<br />
<br />
This session is intended as a contribution to the ontology component of a broader investigation of injury and pain. SNOMED-CT defines an injury as a 'disorder resulting from physical damage to the body'. The WHO defines an injury as a 'bodily lesion at the organic level, resulting from acute exposure to energy (mechanical, thermal, electrical, chemical or radiant), in amounts that exceed the threshold of physiological tolerance.' We will explore these and other definitions with a view to establishing a more coherent understanding of the ontology of injury and of related phenomena such as lesion, trauma, pain, and so forth. Further topics for possible discussion are listed <u>[[Injury | here]]</u>. For the ontology of pain, see in particular [https://philpapers.org/archive/SMITAO-12.pdf here]<br />
<br />
=='''Draft Agenda: Friday Feb 24'''==<br />
<br />
'''08:30-09:15a''' Breakfast and Networking<br />
<br />
'''09:15-10:15a''' Working sessions/Presentations <br />
<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health.<br />
<br />
The ontological representation of occupation presents several challenges. For example, is it your occupation if you have not yet done it? Are you a nuclear engineer before you start your first job doing it? Also, if an occupation is a role (in the BFO sense), then how does it differ from a job role? If I have an occupation of professor, and have had four job roles as professors at different organizations, then that is one occupation numerically but four job roles. How to handle non-employment situations such as "retired" and "homemaker"? What if you concurrently hold two jobs in different "occupations"? I will discuss these issues in the context of recent discussions at the OMRSE monthly meeting and present some options for ontological representation of occupation and related entities.<br />
<br />
'''--break--'''<br />
<br />
'''10:45-12:30p''' Working sessions/Presentations <br />
<br />
*'''Justin Reese:''' Unsupervised machine learning to define subtypes of long COVID using the Human Phenotype Ontology. [[Authors | Co-authors]] <br />
<br />
Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterized by a wide range of manifestations that are difficult to analyze computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.<br />
<br />
We present a method for defining subtypes of long COVID by computationally modeling PASC phenotype data and applying unsupervised machine learning. We extracted Long COVID phenotype data expressed as HPO (Human Phenotype Ontology) terms from electronic healthcare records (EHRs), and then used semantic similarity of phenotype data to calculate a matrix of pairwise patient similarity. This matrix was then used to clustered patients using unsupervised machine learning (k means clustering).<br />
<br />
We identified six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centers to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems.<br />
<br />
Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.<br />
<br />
*'''Paul Heider:''' Beyond Structure: Using Standardized Labels from a Clinical Section Ontology in Natural Language Processing.<br />
<br />
The structured subdivision of clinical notes into sections is often treated as the primary and ultimate use for automatically tagging section boundaries. We discuss how section information (especially when normalized to a standard ontology) can be used to augment natural language processing algorithms, simplify information retrieval tasks,and speed up system development through disaggregated performance evaluation. Our examples emphasize a Clinical Section Ontology recently developed at MUSC but are applicable to any standardized section ontology. We also include a brief overview of the annotated datasets in this domain.<br />
<br />
'''12:30-01:30p''' Networking and Working Lunch <br />
<br />
'''1:00-03:15p''' Working sessions/Presentations from industry partners<br />
<br />
*'''Andy Lin: (VP Strategy, CTO | Mark III Systems)''' NVIDIA DGX H100 – The Premier AI Platform for Research and LLMs.<br />
<br />
With up to 4x training and up to 30x inferencing performance improvements for Large Language Models (LLMs) vs the already industry leading A100 GPU, the recently launched H100 GPU (Hopper) and its purpose-built DGX H100 AI platform are transforming what’s possible with Research and LLMs. In this session, we’ll dive into the DGX H100 platform and how you can get started or accelerate your work with NVIDIA and Mark III.<br />
<br />
*'''John Doole: (TriNetX, LLC)''' Medication Informatics at TriNetX: Powering Clinical Research with a Federated Real World Data Network.<br />
<br />
In this talk we will highlight the importance of medication informatics in clinical research, particularly in the context of a federated network. The talk discusses the challenges of data standardization and interoperability and the benefits of using medication ontologies and classification systems to improve data quality and efficiency. Additionally, the talk presents TriNetX's medication framework and how natural language processing is used to extract medication details, showcasing examples of its implementation and a discussion of the advantages and disadvantages of using NLP in clinical research.<br />
<br />
'''03:15-3:30p''' Closing Remarks<br />
<br />
== '''Participants''' ==<br />
<br />
* Alekseyenko, Alex, Medical University of South Carolina<br />
* Baker, Hamilton, Medical University of South Carolina<br />
* Doole, John, TriNetX<br />
* Heider, Paul, Medical University of South Carolina<br />
* Landgrebe, Jobst, Cognotekt, Cologne<br />
* Hogan, William, co-organizer, University of Florida<br />
* Hutchinson, Tom, University of Pennsylvania, Institute for BioInformatics<br />
* Masci, Anna Maria, National Institute of Environmental Health Sciences (NIEHS)<br />
* Obeid, Jihad, co-organizer, Medical University of South Carolina<br />
* Ohrbach, Richard, University at Buffalo<br />
* Reese, Justin, Lawrence Berkeley National Laboratory<br />
* Scheuermann, Richard, J. Craig Venter Institute<br />
* Simpson, Kit, Medical University of South Carolina<br />
* Skowronek, Matt, TriNetX<br />
* Smith, Barry, co-organizer, University at Buffalo<br />
* Topaloglu, Umit, Center for Biomedical Informatics and Information Technology (CBIIT) National Cancer Institute<br />
* Yonghui Wu, University of Florida</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73414CTS Ontology Workshop 20232023-02-22T14:38:00Z<p>Jihad: /* Draft Agenda: Thursday Feb 23 */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
[[File:Chs23sm.jpg|link=|right|alt=Charleston]]<br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
* Social Determinants of Health<br />
* Mental health<br />
* ChatGPT<br />
<br />
== '''Organizers''' ==<br />
<br />
'''Workshop Co-organizers'''<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
'''CTSOG Co-chairs'''<br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* '''Medical University of South Carolina, the Biomedical Informatics Center''' and:<br />
<br />
[[File:Markiii-logo_160.png|link=https://www.markiiisys.com/|Mark III Systems]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:Nvidia-logo_180.png|link=https://www.nvidia.com/en-us/|NVIDIA]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:TriNetX-Logo-Horizontal-White_180.jpg|link=http://trinetx.com/|TriNetX]]<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Draft Agenda: Thursday Feb 23'''==<br />
<br />
'''08:30a''': Breakfast and Networking<br />
<br />
'''09:00a''': Introduction to the meeting<br />
<br />
'''09:15-10:45a''' Working sessions/Presentations<br />
<br />
*'''Barry Smith and Jobst Landgrebe:''' Discussion on AI and Medicine, with a special focus on ChatGPT<br />
<br />
How are we to explain the peculiar tendency of ChatGPT to throw up what are called [https://buffalo.box.com/v/ChatGPT-Hallucinations ‘hallucinations’]? To answer this question we will draw on our book ''[https://buffalo.app.box.com/v/AI-Without-Fear Why Machines Will Never Rule the World]'', whose core thesis is that an artificial intelligence that could equal or exceed human intelligence—sometimes called "artificial general intelligence" (AGI)—is for mathematical reasons impossible. The argument for this thesis rests on the fact that (1.) human intelligence is a capability of a complex dynamic system—the human brain and central nervous system, and (2.) systems of this sort (like all organic systems) cannot be modelled mathematically in a way that would allow the models to operate inside a computer. We survey on this basis the potential of AI in the future of clinical and translational research. <br />
<br />
'''--break--'''<br />
<br />
'''11:00-12:30p''' Working sessions/Presentations<br />
<br />
*'''Discussion and brief presentations. Topics to include:<br />
:The Clinical and Translational Science ontology landscape in 2023<br />
:Opportunities for AI in medicine<br />
:Will AI make ontologies redundant?<br />
:TBD<br />
<br />
'''12:30-01:30p''' Networking and Working Lunch <br />
<br />
'''01:30-3:00p''' Working sessions/Presentations<br />
<br />
*'''Yonghui Wu:''' A large language model for electronic health records.<br />
<br />
Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
'''--break--'''<br />
<br />
'''3:15-5:00p''' Working sessions/Presentations<br />
<br />
*'''Richard Ohrbach and Barry Smith:''' Defining 'Injury'<br />
<br />
This session is intended as a contribution to the ontology component of a broader investigation of injury and pain. SNOMED-CT defines an injury as a 'disorder resulting from physical damage to the body'. The WHO defines an injury as a 'bodily lesion at the organic level, resulting from acute exposure to energy (mechanical, thermal, electrical, chemical or radiant), in amounts that exceed the threshold of physiological tolerance.' We will explore these and other definitions with a view to establishing a more coherent understanding of the ontology of injury and of related phenomena such as lesion, trauma, pain, and so forth. Further topics for possible discussion are listed <u>[[Injury | here]]</u>. For the ontology of pain, see in particular [https://philpapers.org/archive/SMITAO-12.pdf here]<br />
<br />
=='''Draft Agenda: Friday Feb 24'''==<br />
<br />
09:00-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health.<br />
<br />
The ontological representation of occupation presents several challenges. For example, is it your occupation if you have not yet done it? Are you a nuclear engineer before you start your first job doing it? Also, if an occupation is a role (in the BFO sense), then how does it differ from a job role? If I have an occupation of professor, and have had four job roles as professors at different organizations, then that is one occupation numerically but four job roles. How to handle non-employment situations such as "retired" and "homemaker"? What if you concurrently hold two jobs in different "occupations"? I will discuss these issues in the context of recent discussions at the OMRSE monthly meeting and present some options for ontological representation of occupation and related entities.<br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
*'''Justin Reese:''' Unsupervised machine learning to define subtypes of long COVID using the Human Phenotype Ontology. [[Authors | Co-authors]] <br />
<br />
Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterized by a wide range of manifestations that are difficult to analyze computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.<br />
<br />
We present a method for defining subtypes of long COVID by computationally modeling PASC phenotype data and applying unsupervised machine learning. We extracted Long COVID phenotype data expressed as HPO (Human Phenotype Ontology) terms from electronic healthcare records (EHRs), and then used semantic similarity of phenotype data to calculate a matrix of pairwise patient similarity. This matrix was then used to clustered patients using unsupervised machine learning (k means clustering).<br />
<br />
We identified six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centers to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems.<br />
<br />
Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.<br />
<br />
12:30-01:30p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations from industry partners<br />
<br />
*'''Andy Lin: (VP Strategy, CTO | Mark III Systems)''' NVIDIA DGX H100 – The Premier AI Platform for Research and LLMs.<br />
<br />
With up to 4x training and up to 30x inferencing performance improvements for Large Language Models (LLMs) vs the already industry leading A100 GPU, the recently launched H100 GPU (Hopper) and its purpose-built DGX H100 AI platform are transforming what’s possible with Research and LLMs. In this session, we’ll dive into the DGX H100 platform and how you can get started or accelerate your work with NVIDIA and Mark III.<br />
<br />
*'''John Doole: (TriNetX, LLC)''' Medication Informatics at TriNetX: Powering Clinical Research with a Federated Real World Data Network.<br />
<br />
In this talk we will highlight the importance of medication informatics in clinical research, particularly in the context of a federated network. The talk discusses the challenges of data standardization and interoperability and the benefits of using medication ontologies and classification systems to improve data quality and efficiency. Additionally, the talk presents TriNetX's medication framework and how natural language processing is used to extract medication details, showcasing examples of its implementation and a discussion of the advantages and disadvantages of using NLP in clinical research.<br />
<br />
03:15-'''3:30p''' Closing Remarks<br />
<br />
== '''Participants''' ==<br />
<br />
* Alekseyenko, Alex, Medical University of South Carolina<br />
* Baker, Hamilton, Medical University of South Carolina<br />
* Doole, John, TriNetX<br />
* Heider, Paul, Medical University of South Carolina<br />
* Landgrebe, Jobst, Cognotekt, Cologne<br />
* Hogan, William, co-organizer, University of Florida<br />
* Hutchinson, Tom, University of Pennsylvania, Institute for BioInformatics<br />
* Masci, Anna Maria, National Institute of Environmental Health Sciences (NIEHS)<br />
* Obeid, Jihad, co-organizer, Medical University of South Carolina<br />
* Ohrbach, Richard, University at Buffalo<br />
* Reese, Justin, Lawrence Berkeley National Laboratory<br />
* Scheuermann, Richard, J. Craig Venter Institute<br />
* Simpson, Kit, Medical University of South Carolina<br />
* Skowronek, Matt, TriNetX<br />
* Smith, Barry, co-organizer, University at Buffalo<br />
* Topaloglu, Umit, Center for Biomedical Informatics and Information Technology (CBIIT) National Cancer Institute<br />
* Yonghui Wu, University of Florida</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73413CTS Ontology Workshop 20232023-02-21T18:01:03Z<p>Jihad: /* Draft Agenda: Friday Feb 24 */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
[[File:Chs23sm.jpg|link=|right|alt=Charleston]]<br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
* Social Determinants of Health<br />
* Mental health<br />
* ChatGPT<br />
<br />
== '''Organizers''' ==<br />
<br />
'''Workshop Co-organizers'''<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
'''CTSOG Co-chairs'''<br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* '''Medical University of South Carolina, the Biomedical Informatics Center''' and:<br />
<br />
[[File:Markiii-logo_160.png|link=https://www.markiiisys.com/|Mark III Systems]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:Nvidia-logo_180.png|link=https://www.nvidia.com/en-us/|NVIDIA]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:TriNetX-Logo-Horizontal-White_180.jpg|link=http://trinetx.com/|TriNetX]]<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Draft Agenda: Thursday Feb 23'''==<br />
<br />
08:30a: Breakfast and Networking<br />
<br />
09:00a: Introduction to the meeting<br />
<br />
09:15-11:00a Working sessions/Presentations<br />
<br />
*'''Barry Smith and Jobst Landgrebe:''' Discussion on AI and Medicine, with a special focus on ChatGPT<br />
<br />
How are we to explain the peculiar tendency of ChatGPT to throw up what are called [https://buffalo.box.com/v/ChatGPT-Hallucinations ‘hallucinations’]? To answer this question we will draw on our book ''[https://buffalo.app.box.com/v/AI-Without-Fear Why Machines Will Never Rule the World]'', whose core thesis is that an artificial intelligence that could equal or exceed human intelligence—sometimes called "artificial general intelligence" (AGI)—is for mathematical reasons impossible. The argument for this thesis rests on the fact that (1.) human intelligence is a capability of a complex dynamic system—the human brain and central nervous system, and (2.) systems of this sort (like all organic systems) cannot be modelled mathematically in a way that would allow the models to operate inside a computer. We survey on this basis the potential of AI in the future of clinical and translational research. <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations<br />
<br />
*'''Discussion and brief presentations. Topics to include:<br />
:The Clinical and Translational Science ontology landscape in 2023<br />
:Opportunities for AI in medicine<br />
:Will AI make ontologies redundant?<br />
:TBD<br />
<br />
12:30-01:30p Networking and Working Lunch <br />
<br />
01:30-3:00p Working sessions/Presentations<br />
<br />
*'''Yonghui Wu:''' A large language model for electronic health records.<br />
<br />
Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
--break--<br />
<br />
3:15-5:00p Working sessions/Presentations<br />
<br />
*'''Richard Ohrbach and Barry Smith:''' Defining 'Injury'<br />
<br />
This session is intended as a contribution to the ontology component of a broader investigation of injury and pain. SNOMED-CT defines an injury as a 'disorder resulting from physical damage to the body'. The WHO defines an injury as a 'bodily lesion at the organic level, resulting from acute exposure to energy (mechanical, thermal, electrical, chemical or radiant), in amounts that exceed the threshold of physiological tolerance.' We will explore these and other definitions with a view to establishing a more coherent understanding of the ontology of injury and of related phenomena such as lesion, trauma, pain, and so forth. Further topics for possible discussion are listed <u>[[Injury | here]]</u>. For the ontology of pain, see in particular [https://philpapers.org/archive/SMITAO-12.pdf here]<br />
<br />
=='''Draft Agenda: Friday Feb 24'''==<br />
<br />
09:00-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health.<br />
<br />
The ontological representation of occupation presents several challenges. For example, is it your occupation if you have not yet done it? Are you a nuclear engineer before you start your first job doing it? Also, if an occupation is a role (in the BFO sense), then how does it differ from a job role? If I have an occupation of professor, and have had four job roles as professors at different organizations, then that is one occupation numerically but four job roles. How to handle non-employment situations such as "retired" and "homemaker"? What if you concurrently hold two jobs in different "occupations"? I will discuss these issues in the context of recent discussions at the OMRSE monthly meeting and present some options for ontological representation of occupation and related entities.<br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
*'''Justin Reese:''' Unsupervised machine learning to define subtypes of long COVID using the Human Phenotype Ontology. [[Authors | Co-authors]] <br />
<br />
Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterized by a wide range of manifestations that are difficult to analyze computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.<br />
<br />
We present a method for defining subtypes of long COVID by computationally modeling PASC phenotype data and applying unsupervised machine learning. We extracted Long COVID phenotype data expressed as HPO (Human Phenotype Ontology) terms from electronic healthcare records (EHRs), and then used semantic similarity of phenotype data to calculate a matrix of pairwise patient similarity. This matrix was then used to clustered patients using unsupervised machine learning (k means clustering).<br />
<br />
We identified six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centers to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems.<br />
<br />
Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.<br />
<br />
12:30-01:30p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations from industry partners<br />
<br />
*'''Andy Lin: (VP Strategy, CTO | Mark III Systems)''' NVIDIA DGX H100 – The Premier AI Platform for Research and LLMs.<br />
<br />
With up to 4x training and up to 30x inferencing performance improvements for Large Language Models (LLMs) vs the already industry leading A100 GPU, the recently launched H100 GPU (Hopper) and its purpose-built DGX H100 AI platform are transforming what’s possible with Research and LLMs. In this session, we’ll dive into the DGX H100 platform and how you can get started or accelerate your work with NVIDIA and Mark III.<br />
<br />
*'''John Doole: (TriNetX, LLC)''' Medication Informatics at TriNetX: Powering Clinical Research with a Federated Real World Data Network.<br />
<br />
In this talk we will highlight the importance of medication informatics in clinical research, particularly in the context of a federated network. The talk discusses the challenges of data standardization and interoperability and the benefits of using medication ontologies and classification systems to improve data quality and efficiency. Additionally, the talk presents TriNetX's medication framework and how natural language processing is used to extract medication details, showcasing examples of its implementation and a discussion of the advantages and disadvantages of using NLP in clinical research.<br />
<br />
03:15-'''3:30p''' Closing Remarks<br />
<br />
== '''Participants''' ==<br />
<br />
* Alekseyenko, Alex, Medical University of South Carolina<br />
* Baker, Hamilton, Medical University of South Carolina<br />
* Doole, John, TriNetX<br />
* Heider, Paul, Medical University of South Carolina<br />
* Landgrebe, Jobst, Cognotekt, Cologne<br />
* Hogan, William, co-organizer, University of Florida<br />
* Hutchinson, Tom, University of Pennsylvania, Institute for BioInformatics<br />
* Masci, Anna Maria, National Institute of Environmental Health Sciences (NIEHS)<br />
* Obeid, Jihad, co-organizer, Medical University of South Carolina<br />
* Ohrbach, Richard, University at Buffalo<br />
* Reese, Justin, Lawrence Berkeley National Laboratory<br />
* Scheuermann, Richard, J. Craig Venter Institute<br />
* Simpson, Kit, Medical University of South Carolina<br />
* Skowronek, Matt, TriNetX<br />
* Smith, Barry, co-organizer, University at Buffalo<br />
* Topaloglu, Umit, Center for Biomedical Informatics and Information Technology (CBIIT) National Cancer Institute<br />
* Yonghui Wu, University of Florida</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73412CTS Ontology Workshop 20232023-02-21T18:00:47Z<p>Jihad: /* Draft Agenda: Thursday Feb 23 */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
[[File:Chs23sm.jpg|link=|right|alt=Charleston]]<br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
* Social Determinants of Health<br />
* Mental health<br />
* ChatGPT<br />
<br />
== '''Organizers''' ==<br />
<br />
'''Workshop Co-organizers'''<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
'''CTSOG Co-chairs'''<br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* '''Medical University of South Carolina, the Biomedical Informatics Center''' and:<br />
<br />
[[File:Markiii-logo_160.png|link=https://www.markiiisys.com/|Mark III Systems]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:Nvidia-logo_180.png|link=https://www.nvidia.com/en-us/|NVIDIA]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:TriNetX-Logo-Horizontal-White_180.jpg|link=http://trinetx.com/|TriNetX]]<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Draft Agenda: Thursday Feb 23'''==<br />
<br />
08:30a: Breakfast and Networking<br />
<br />
09:00a: Introduction to the meeting<br />
<br />
09:15-11:00a Working sessions/Presentations<br />
<br />
*'''Barry Smith and Jobst Landgrebe:''' Discussion on AI and Medicine, with a special focus on ChatGPT<br />
<br />
How are we to explain the peculiar tendency of ChatGPT to throw up what are called [https://buffalo.box.com/v/ChatGPT-Hallucinations ‘hallucinations’]? To answer this question we will draw on our book ''[https://buffalo.app.box.com/v/AI-Without-Fear Why Machines Will Never Rule the World]'', whose core thesis is that an artificial intelligence that could equal or exceed human intelligence—sometimes called "artificial general intelligence" (AGI)—is for mathematical reasons impossible. The argument for this thesis rests on the fact that (1.) human intelligence is a capability of a complex dynamic system—the human brain and central nervous system, and (2.) systems of this sort (like all organic systems) cannot be modelled mathematically in a way that would allow the models to operate inside a computer. We survey on this basis the potential of AI in the future of clinical and translational research. <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations<br />
<br />
*'''Discussion and brief presentations. Topics to include:<br />
:The Clinical and Translational Science ontology landscape in 2023<br />
:Opportunities for AI in medicine<br />
:Will AI make ontologies redundant?<br />
:TBD<br />
<br />
12:30-01:30p Networking and Working Lunch <br />
<br />
01:30-3:00p Working sessions/Presentations<br />
<br />
*'''Yonghui Wu:''' A large language model for electronic health records.<br />
<br />
Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
--break--<br />
<br />
3:15-5:00p Working sessions/Presentations<br />
<br />
*'''Richard Ohrbach and Barry Smith:''' Defining 'Injury'<br />
<br />
This session is intended as a contribution to the ontology component of a broader investigation of injury and pain. SNOMED-CT defines an injury as a 'disorder resulting from physical damage to the body'. The WHO defines an injury as a 'bodily lesion at the organic level, resulting from acute exposure to energy (mechanical, thermal, electrical, chemical or radiant), in amounts that exceed the threshold of physiological tolerance.' We will explore these and other definitions with a view to establishing a more coherent understanding of the ontology of injury and of related phenomena such as lesion, trauma, pain, and so forth. Further topics for possible discussion are listed <u>[[Injury | here]]</u>. For the ontology of pain, see in particular [https://philpapers.org/archive/SMITAO-12.pdf here]<br />
<br />
=='''Draft Agenda: Friday Feb 24'''==<br />
<br />
09:00-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health.<br />
<br />
The ontological representation of occupation presents several challenges. For example, is it your occupation if you have not yet done it? Are you a nuclear engineer before you start your first job doing it? Also, if an occupation is a role (in the BFO sense), then how does it differ from a job role? If I have an occupation of professor, and have had four job roles as professors at different organizations, then that is one occupation numerically but four job roles. How to handle non-employment situations such as "retired" and "homemaker"? What if you concurrently hold two jobs in different "occupations"? I will discuss these issues in the context of recent discussions at the OMRSE monthly meeting and present some options for ontological representation of occupation and related entities.<br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
*'''Justin Reese:''' Unsupervised machine learning to define subtypes of long COVID using the Human Phenotype Ontology. [[Authors | Co-authors]] <br />
<br />
'''Abstract:''' Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterized by a wide range of manifestations that are difficult to analyze computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.<br />
<br />
We present a method for defining subtypes of long COVID by computationally modeling PASC phenotype data and applying unsupervised machine learning. We extracted Long COVID phenotype data expressed as HPO (Human Phenotype Ontology) terms from electronic healthcare records (EHRs), and then used semantic similarity of phenotype data to calculate a matrix of pairwise patient similarity. This matrix was then used to clustered patients using unsupervised machine learning (k means clustering).<br />
<br />
We identified six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centers to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems.<br />
<br />
Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.<br />
<br />
12:30-01:30p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations from industry partners<br />
<br />
*'''Andy Lin: (VP Strategy, CTO | Mark III Systems)''' NVIDIA DGX H100 – The Premier AI Platform for Research and LLMs.<br />
<br />
With up to 4x training and up to 30x inferencing performance improvements for Large Language Models (LLMs) vs the already industry leading A100 GPU, the recently launched H100 GPU (Hopper) and its purpose-built DGX H100 AI platform are transforming what’s possible with Research and LLMs. In this session, we’ll dive into the DGX H100 platform and how you can get started or accelerate your work with NVIDIA and Mark III.<br />
<br />
*'''John Doole: (TriNetX, LLC)''' Medication Informatics at TriNetX: Powering Clinical Research with a Federated Real World Data Network.<br />
<br />
In this talk we will highlight the importance of medication informatics in clinical research, particularly in the context of a federated network. The talk discusses the challenges of data standardization and interoperability and the benefits of using medication ontologies and classification systems to improve data quality and efficiency. Additionally, the talk presents TriNetX's medication framework and how natural language processing is used to extract medication details, showcasing examples of its implementation and a discussion of the advantages and disadvantages of using NLP in clinical research.<br />
<br />
03:15-'''3:30p''' Closing Remarks<br />
<br />
== '''Participants''' ==<br />
<br />
* Alekseyenko, Alex, Medical University of South Carolina<br />
* Baker, Hamilton, Medical University of South Carolina<br />
* Doole, John, TriNetX<br />
* Heider, Paul, Medical University of South Carolina<br />
* Landgrebe, Jobst, Cognotekt, Cologne<br />
* Hogan, William, co-organizer, University of Florida<br />
* Hutchinson, Tom, University of Pennsylvania, Institute for BioInformatics<br />
* Masci, Anna Maria, National Institute of Environmental Health Sciences (NIEHS)<br />
* Obeid, Jihad, co-organizer, Medical University of South Carolina<br />
* Ohrbach, Richard, University at Buffalo<br />
* Reese, Justin, Lawrence Berkeley National Laboratory<br />
* Scheuermann, Richard, J. Craig Venter Institute<br />
* Simpson, Kit, Medical University of South Carolina<br />
* Skowronek, Matt, TriNetX<br />
* Smith, Barry, co-organizer, University at Buffalo<br />
* Topaloglu, Umit, Center for Biomedical Informatics and Information Technology (CBIIT) National Cancer Institute<br />
* Yonghui Wu, University of Florida</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73411CTS Ontology Workshop 20232023-02-21T17:57:31Z<p>Jihad: /* Draft Agenda: Friday Feb 24 */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
[[File:Chs23sm.jpg|link=|right|alt=Charleston]]<br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
* Social Determinants of Health<br />
* Mental health<br />
* ChatGPT<br />
<br />
== '''Organizers''' ==<br />
<br />
'''Workshop Co-organizers'''<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
'''CTSOG Co-chairs'''<br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* '''Medical University of South Carolina, the Biomedical Informatics Center''' and:<br />
<br />
[[File:Markiii-logo_160.png|link=https://www.markiiisys.com/|Mark III Systems]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:Nvidia-logo_180.png|link=https://www.nvidia.com/en-us/|NVIDIA]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:TriNetX-Logo-Horizontal-White_180.jpg|link=http://trinetx.com/|TriNetX]]<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Draft Agenda: Thursday Feb 23'''==<br />
<br />
08:30a: Breakfast and Networking<br />
<br />
09:00a: Introduction to the meeting<br />
<br />
09:15-11:00a Working sessions/Presentations<br />
<br />
*'''Barry Smith and Jobst Landgrebe:''' Discussion on AI and Medicine, with a special focus on ChatGPT<br />
<br />
'''Abstract: '''How are we to explain the peculiar tendency of ChatGPT to throw up what are called [https://buffalo.box.com/v/ChatGPT-Hallucinations ‘hallucinations’]? To answer this question we will draw on our book ''[https://buffalo.app.box.com/v/AI-Without-Fear Why Machines Will Never Rule the World]'', whose core thesis is that an artificial intelligence that could equal or exceed human intelligence—sometimes called "artificial general intelligence" (AGI)—is for mathematical reasons impossible. The argument for this thesis rests on the fact that (1.) human intelligence is a capability of a complex dynamic system—the human brain and central nervous system, and (2.) systems of this sort (like all organic systems) cannot be modelled mathematically in a way that would allow the models to operate inside a computer. We survey on this basis the potential of AI in the future of clinical and translational research. <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations<br />
<br />
*'''Discussion and brief presentations. Topics to include:<br />
:The Clinical and Translational Science ontology landscape in 2023<br />
:Opportunities for AI in medicine<br />
:Will AI make ontologies redundant?<br />
:TBD<br />
<br />
12:30-01:30p Networking and Working Lunch <br />
<br />
01:30-3:00p Working sessions/Presentations<br />
<br />
*'''Yonghui Wu:''' A large language model for electronic health records.<br />
<br />
'''Abstract:''' Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
--break--<br />
<br />
3:15-5:00p Working sessions/Presentations<br />
<br />
*'''Richard Ohrbach and Barry Smith:''' Defining 'Injury'<br />
<br />
'''Abstract:''' This session is intended as a contribution to the ontology component of a broader investigation of injury and pain. SNOMED-CT defines an injury as a 'disorder resulting from physical damage to the body'. The WHO defines an injury as a 'bodily lesion at the organic level, resulting from acute exposure to energy (mechanical, thermal, electrical, chemical or radiant), in amounts that exceed the threshold of physiological tolerance.' We will explore these and other definitions with a view to establishing a more coherent understanding of the ontology of injury and of related phenomena such as lesion, trauma, pain, and so forth. Further topics for possible discussion are listed <u>[[Injury | here]]</u>. For the ontology of pain, see in particular [https://philpapers.org/archive/SMITAO-12.pdf here]<br />
<br />
=='''Draft Agenda: Friday Feb 24'''==<br />
<br />
09:00-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health.<br />
<br />
The ontological representation of occupation presents several challenges. For example, is it your occupation if you have not yet done it? Are you a nuclear engineer before you start your first job doing it? Also, if an occupation is a role (in the BFO sense), then how does it differ from a job role? If I have an occupation of professor, and have had four job roles as professors at different organizations, then that is one occupation numerically but four job roles. How to handle non-employment situations such as "retired" and "homemaker"? What if you concurrently hold two jobs in different "occupations"? I will discuss these issues in the context of recent discussions at the OMRSE monthly meeting and present some options for ontological representation of occupation and related entities.<br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
*'''Justin Reese:''' Unsupervised machine learning to define subtypes of long COVID using the Human Phenotype Ontology. [[Authors | Co-authors]] <br />
<br />
'''Abstract:''' Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterized by a wide range of manifestations that are difficult to analyze computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.<br />
<br />
We present a method for defining subtypes of long COVID by computationally modeling PASC phenotype data and applying unsupervised machine learning. We extracted Long COVID phenotype data expressed as HPO (Human Phenotype Ontology) terms from electronic healthcare records (EHRs), and then used semantic similarity of phenotype data to calculate a matrix of pairwise patient similarity. This matrix was then used to clustered patients using unsupervised machine learning (k means clustering).<br />
<br />
We identified six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centers to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems.<br />
<br />
Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.<br />
<br />
12:30-01:30p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations from industry partners<br />
<br />
*'''Andy Lin: (VP Strategy, CTO | Mark III Systems)''' NVIDIA DGX H100 – The Premier AI Platform for Research and LLMs.<br />
<br />
With up to 4x training and up to 30x inferencing performance improvements for Large Language Models (LLMs) vs the already industry leading A100 GPU, the recently launched H100 GPU (Hopper) and its purpose-built DGX H100 AI platform are transforming what’s possible with Research and LLMs. In this session, we’ll dive into the DGX H100 platform and how you can get started or accelerate your work with NVIDIA and Mark III.<br />
<br />
*'''John Doole: (TriNetX, LLC)''' Medication Informatics at TriNetX: Powering Clinical Research with a Federated Real World Data Network.<br />
<br />
In this talk we will highlight the importance of medication informatics in clinical research, particularly in the context of a federated network. The talk discusses the challenges of data standardization and interoperability and the benefits of using medication ontologies and classification systems to improve data quality and efficiency. Additionally, the talk presents TriNetX's medication framework and how natural language processing is used to extract medication details, showcasing examples of its implementation and a discussion of the advantages and disadvantages of using NLP in clinical research.<br />
<br />
03:15-'''3:30p''' Closing Remarks<br />
<br />
== '''Participants''' ==<br />
<br />
* Alekseyenko, Alex, Medical University of South Carolina<br />
* Baker, Hamilton, Medical University of South Carolina<br />
* Doole, John, TriNetX<br />
* Heider, Paul, Medical University of South Carolina<br />
* Landgrebe, Jobst, Cognotekt, Cologne<br />
* Hogan, William, co-organizer, University of Florida<br />
* Hutchinson, Tom, University of Pennsylvania, Institute for BioInformatics<br />
* Masci, Anna Maria, National Institute of Environmental Health Sciences (NIEHS)<br />
* Obeid, Jihad, co-organizer, Medical University of South Carolina<br />
* Ohrbach, Richard, University at Buffalo<br />
* Reese, Justin, Lawrence Berkeley National Laboratory<br />
* Scheuermann, Richard, J. Craig Venter Institute<br />
* Simpson, Kit, Medical University of South Carolina<br />
* Skowronek, Matt, TriNetX<br />
* Smith, Barry, co-organizer, University at Buffalo<br />
* Topaloglu, Umit, Center for Biomedical Informatics and Information Technology (CBIIT) National Cancer Institute<br />
* Yonghui Wu, University of Florida</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73410CTS Ontology Workshop 20232023-02-21T16:04:22Z<p>Jihad: /* Draft Agenda: Friday Feb 24 */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
[[File:Chs23sm.jpg|link=|right|alt=Charleston]]<br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
* Social Determinants of Health<br />
* Mental health<br />
* ChatGPT<br />
<br />
== '''Organizers''' ==<br />
<br />
'''Workshop Co-organizers'''<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
'''CTSOG Co-chairs'''<br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* '''Medical University of South Carolina, the Biomedical Informatics Center''' and:<br />
<br />
[[File:Markiii-logo_160.png|link=https://www.markiiisys.com/|Mark III Systems]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:Nvidia-logo_180.png|link=https://www.nvidia.com/en-us/|NVIDIA]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:TriNetX-Logo-Horizontal-White_180.jpg|link=http://trinetx.com/|TriNetX]]<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Draft Agenda: Thursday Feb 23'''==<br />
<br />
08:30a: Breakfast and Networking<br />
<br />
09:00a: Introduction to the meeting<br />
<br />
09:15-11:00a Working sessions/Presentations<br />
<br />
*'''Barry Smith and Jobst Landgrebe:''' Discussion on AI and Medicine, with a special focus on ChatGPT<br />
<br />
'''Abstract: '''How are we to explain the peculiar tendency of ChatGPT to throw up what are called [https://buffalo.box.com/v/ChatGPT-Hallucinations ‘hallucinations’]? To answer this question we will draw on our book ''[https://buffalo.app.box.com/v/AI-Without-Fear Why Machines Will Never Rule the World]'', whose core thesis is that an artificial intelligence that could equal or exceed human intelligence—sometimes called "artificial general intelligence" (AGI)—is for mathematical reasons impossible. The argument for this thesis rests on the fact that (1.) human intelligence is a capability of a complex dynamic system—the human brain and central nervous system, and (2.) systems of this sort (like all organic systems) cannot be modelled mathematically in a way that would allow the models to operate inside a computer. We survey on this basis the potential of AI in the future of clinical and translational research. <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations<br />
<br />
*'''Discussion and brief presentations. Topics to include:<br />
:The Clinical and Translational Science ontology landscape in 2023<br />
:Opportunities for AI in medicine<br />
:Will AI make ontologies redundant?<br />
:TBD<br />
<br />
12:30-01:30p Networking and Working Lunch <br />
<br />
01:30-3:00p Working sessions/Presentations<br />
<br />
*'''Yonghui Wu:''' A large language model for electronic health records.<br />
<br />
'''Abstract:''' Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
--break--<br />
<br />
3:15-5:00p Working sessions/Presentations<br />
<br />
*'''Richard Ohrbach and Barry Smith:''' Defining 'Injury'<br />
<br />
'''Abstract:''' This session is intended as a contribution to the ontology component of a broader investigation of injury and pain. SNOMED-CT defines an injury as a 'disorder resulting from physical damage to the body'. The WHO defines an injury as a 'bodily lesion at the organic level, resulting from acute exposure to energy (mechanical, thermal, electrical, chemical or radiant), in amounts that exceed the threshold of physiological tolerance.' We will explore these and other definitions with a view to establishing a more coherent understanding of the ontology of injury and of related phenomena such as lesion, trauma, pain, and so forth. Further topics for possible discussion are listed <u>[[Injury | here]]</u>. For the ontology of pain, see in particular [https://philpapers.org/archive/SMITAO-12.pdf here]<br />
<br />
=='''Draft Agenda: Friday Feb 24'''==<br />
<br />
09:00-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health.<br />
<br />
The ontological representation of occupation presents several challenges. For example, is it your occupation if you have not yet done it? Are you a nuclear engineer before you start your first job doing it? Also, if an occupation is a role (in the BFO sense), then how does it differ from a job role? If I have an occupation of professor, and have had four job roles as professors at different organizations, then that is one occupation numerically but four job roles. How to handle non-employment situations such as "retired" and "homemaker"? What if you concurrently hold two jobs in different "occupations"? I will discuss these issues in the context of recent discussions at the OMRSE monthly meeting and present some options for ontological representation of occupation and related entities.<br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
*'''Justin Reese:''' Unsupervised machine learning to define subtypes of long COVID using the Human Phenotype Ontology. [[Authors | Co-authors]] <br />
<br />
'''Abstract:''' Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterized by a wide range of manifestations that are difficult to analyze computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.<br />
<br />
We present a method for defining subtypes of long COVID by computationally modeling PASC phenotype data and applying unsupervised machine learning. We extracted Long COVID phenotype data expressed as HPO (Human Phenotype Ontology) terms from electronic healthcare records (EHRs), and then used semantic similarity of phenotype data to calculate a matrix of pairwise patient similarity. This matrix was then used to clustered patients using unsupervised machine learning (k means clustering).<br />
<br />
We identified six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centers to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems.<br />
<br />
Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.<br />
<br />
12:30-01:30p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations from industry partners<br />
<br />
*'''Andy Lin: (VP Strategy, CTO | Mark III Systems)''' NVIDIA DGX H100 – The Premier AI Platform for Research and LLMs.<br />
<br />
With up to 4x training and up to 30x inferencing performance improvements for Large Language Models (LLMs) vs the already industry leading A100 GPU, the recently launched H100 GPU (Hopper) and its purpose-built DGX H100 AI platform are transforming what’s possible with Research and LLMs. In this session, we’ll dive into the DGX H100 platform and how you can get started or accelerate your work with NVIDIA and Mark III.<br />
<br />
*'''John Doole: (TriNetX, LLC)''' .<br />
<br />
.<br />
<br />
<br />
03:15-'''3:30p''' Closing Remarks<br />
<br />
== '''Participants''' ==<br />
<br />
* Alekseyenko, Alex, Medical University of South Carolina<br />
* Baker, Hamilton, Medical University of South Carolina<br />
* Doole, John, TriNetX<br />
* Heider, Paul, Medical University of South Carolina<br />
* Landgrebe, Jobst, Cognotekt, Cologne<br />
* Hogan, William, co-organizer, University of Florida<br />
* Hutchinson, Tom, University of Pennsylvania, Institute for BioInformatics<br />
* Masci, Anna Maria, National Institute of Environmental Health Sciences (NIEHS)<br />
* Obeid, Jihad, co-organizer, Medical University of South Carolina<br />
* Ohrbach, Richard, University at Buffalo<br />
* Reese, Justin, Lawrence Berkeley National Laboratory<br />
* Scheuermann, Richard, J. Craig Venter Institute<br />
* Simpson, Kit, Medical University of South Carolina<br />
* Skowronek, Matt, TriNetX<br />
* Smith, Barry, co-organizer, University at Buffalo<br />
* Topaloglu, Umit, Center for Biomedical Informatics and Information Technology (CBIIT) National Cancer Institute<br />
* Yonghui Wu, University of Florida</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73409CTS Ontology Workshop 20232023-02-21T15:42:43Z<p>Jihad: /* Draft Agenda: Friday Feb 24 */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
[[File:Chs23sm.jpg|link=|right|alt=Charleston]]<br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
* Social Determinants of Health<br />
* Mental health<br />
* ChatGPT<br />
<br />
== '''Organizers''' ==<br />
<br />
'''Workshop Co-organizers'''<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
'''CTSOG Co-chairs'''<br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* '''Medical University of South Carolina, the Biomedical Informatics Center''' and:<br />
<br />
[[File:Markiii-logo_160.png|link=https://www.markiiisys.com/|Mark III Systems]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:Nvidia-logo_180.png|link=https://www.nvidia.com/en-us/|NVIDIA]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:TriNetX-Logo-Horizontal-White_180.jpg|link=http://trinetx.com/|TriNetX]]<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Draft Agenda: Thursday Feb 23'''==<br />
<br />
08:30a: Breakfast and Networking<br />
<br />
09:00a: Introduction to the meeting<br />
<br />
09:15-11:00a Working sessions/Presentations<br />
<br />
*'''Barry Smith and Jobst Landgrebe:''' Discussion on AI and Medicine, with a special focus on ChatGPT<br />
<br />
'''Abstract: '''How are we to explain the peculiar tendency of ChatGPT to throw up what are called [https://buffalo.box.com/v/ChatGPT-Hallucinations ‘hallucinations’]? To answer this question we will draw on our book ''[https://buffalo.app.box.com/v/AI-Without-Fear Why Machines Will Never Rule the World]'', whose core thesis is that an artificial intelligence that could equal or exceed human intelligence—sometimes called "artificial general intelligence" (AGI)—is for mathematical reasons impossible. The argument for this thesis rests on the fact that (1.) human intelligence is a capability of a complex dynamic system—the human brain and central nervous system, and (2.) systems of this sort (like all organic systems) cannot be modelled mathematically in a way that would allow the models to operate inside a computer. We survey on this basis the potential of AI in the future of clinical and translational research. <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations<br />
<br />
*'''Discussion and brief presentations. Topics to include:<br />
:The Clinical and Translational Science ontology landscape in 2023<br />
:Opportunities for AI in medicine<br />
:Will AI make ontologies redundant?<br />
:TBD<br />
<br />
12:30-01:30p Networking and Working Lunch <br />
<br />
01:30-3:00p Working sessions/Presentations<br />
<br />
*'''Yonghui Wu:''' A large language model for electronic health records.<br />
<br />
'''Abstract:''' Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
--break--<br />
<br />
3:15-5:00p Working sessions/Presentations<br />
<br />
*'''Richard Ohrbach and Barry Smith:''' Defining 'Injury'<br />
<br />
'''Abstract:''' This session is intended as a contribution to the ontology component of a broader investigation of injury and pain. SNOMED-CT defines an injury as a 'disorder resulting from physical damage to the body'. The WHO defines an injury as a 'bodily lesion at the organic level, resulting from acute exposure to energy (mechanical, thermal, electrical, chemical or radiant), in amounts that exceed the threshold of physiological tolerance.' We will explore these and other definitions with a view to establishing a more coherent understanding of the ontology of injury and of related phenomena such as lesion, trauma, pain, and so forth. Further topics for possible discussion are listed <u>[[Injury | here]]</u>. For the ontology of pain, see in particular [https://philpapers.org/archive/SMITAO-12.pdf here]<br />
<br />
=='''Draft Agenda: Friday Feb 24'''==<br />
<br />
09:00-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health.<br />
<br />
The ontological representation of occupation presents several challenges. For example, is it your occupation if you have not yet done it? Are you a nuclear engineer before you start your first job doing it? Also, if an occupation is a role (in the BFO sense), then how does it differ from a job role? If I have an occupation of professor, and have had four job roles as professors at different organizations, then that is one occupation numerically but four job roles. How to handle non-employment situations such as "retired" and "homemaker"? What if you concurrently hold two jobs in different "occupations"? I will discuss these issues in the context of recent discussions at the OMRSE monthly meeting and present some options for ontological representation of occupation and related entities.<br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
*'''Justin Reese:''' Unsupervised machine learning to define subtypes of long COVID using the Human Phenotype Ontology. [[Authors | Co-authors]] <br />
<br />
'''Abstract:''' Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterized by a wide range of manifestations that are difficult to analyze computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.<br />
<br />
We present a method for defining subtypes of long COVID by computationally modeling PASC phenotype data and applying unsupervised machine learning. We extracted Long COVID phenotype data expressed as HPO (Human Phenotype Ontology) terms from electronic healthcare records (EHRs), and then used semantic similarity of phenotype data to calculate a matrix of pairwise patient similarity. This matrix was then used to clustered patients using unsupervised machine learning (k means clustering).<br />
<br />
We identified six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centers to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems.<br />
<br />
Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.<br />
<br />
12:30-01:30p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations from industry partners<br />
<br />
TBD<br />
<br />
03:15-'''3:30p''' Closing Remarks<br />
<br />
== '''Participants''' ==<br />
<br />
* Alekseyenko, Alex, Medical University of South Carolina<br />
* Baker, Hamilton, Medical University of South Carolina<br />
* Doole, John, TriNetX<br />
* Heider, Paul, Medical University of South Carolina<br />
* Landgrebe, Jobst, Cognotekt, Cologne<br />
* Hogan, William, co-organizer, University of Florida<br />
* Hutchinson, Tom, University of Pennsylvania, Institute for BioInformatics<br />
* Masci, Anna Maria, National Institute of Environmental Health Sciences (NIEHS)<br />
* Obeid, Jihad, co-organizer, Medical University of South Carolina<br />
* Ohrbach, Richard, University at Buffalo<br />
* Reese, Justin, Lawrence Berkeley National Laboratory<br />
* Scheuermann, Richard, J. Craig Venter Institute<br />
* Simpson, Kit, Medical University of South Carolina<br />
* Skowronek, Matt, TriNetX<br />
* Smith, Barry, co-organizer, University at Buffalo<br />
* Topaloglu, Umit, Center for Biomedical Informatics and Information Technology (CBIIT) National Cancer Institute<br />
* Yonghui Wu, University of Florida</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73408CTS Ontology Workshop 20232023-02-21T15:40:53Z<p>Jihad: /* Participants */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
[[File:Chs23sm.jpg|link=|right|alt=Charleston]]<br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
* Social Determinants of Health<br />
* Mental health<br />
* ChatGPT<br />
<br />
== '''Organizers''' ==<br />
<br />
'''Workshop Co-organizers'''<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
'''CTSOG Co-chairs'''<br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* '''Medical University of South Carolina, the Biomedical Informatics Center''' and:<br />
<br />
[[File:Markiii-logo_160.png|link=https://www.markiiisys.com/|Mark III Systems]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:Nvidia-logo_180.png|link=https://www.nvidia.com/en-us/|NVIDIA]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:TriNetX-Logo-Horizontal-White_180.jpg|link=http://trinetx.com/|TriNetX]]<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Draft Agenda: Thursday Feb 23'''==<br />
<br />
08:30a: Breakfast and Networking<br />
<br />
09:00a: Introduction to the meeting<br />
<br />
09:15-11:00a Working sessions/Presentations<br />
<br />
*'''Barry Smith and Jobst Landgrebe:''' Discussion on AI and Medicine, with a special focus on ChatGPT<br />
<br />
'''Abstract: '''How are we to explain the peculiar tendency of ChatGPT to throw up what are called [https://buffalo.box.com/v/ChatGPT-Hallucinations ‘hallucinations’]? To answer this question we will draw on our book ''[https://buffalo.app.box.com/v/AI-Without-Fear Why Machines Will Never Rule the World]'', whose core thesis is that an artificial intelligence that could equal or exceed human intelligence—sometimes called "artificial general intelligence" (AGI)—is for mathematical reasons impossible. The argument for this thesis rests on the fact that (1.) human intelligence is a capability of a complex dynamic system—the human brain and central nervous system, and (2.) systems of this sort (like all organic systems) cannot be modelled mathematically in a way that would allow the models to operate inside a computer. We survey on this basis the potential of AI in the future of clinical and translational research. <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations<br />
<br />
*'''Discussion and brief presentations. Topics to include:<br />
:The Clinical and Translational Science ontology landscape in 2023<br />
:Opportunities for AI in medicine<br />
:Will AI make ontologies redundant?<br />
:TBD<br />
<br />
12:30-01:30p Networking and Working Lunch <br />
<br />
01:30-3:00p Working sessions/Presentations<br />
<br />
*'''Yonghui Wu:''' A large language model for electronic health records.<br />
<br />
'''Abstract:''' Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
--break--<br />
<br />
3:15-5:00p Working sessions/Presentations<br />
<br />
*'''Richard Ohrbach and Barry Smith:''' Defining 'Injury'<br />
<br />
'''Abstract:''' This session is intended as a contribution to the ontology component of a broader investigation of injury and pain. SNOMED-CT defines an injury as a 'disorder resulting from physical damage to the body'. The WHO defines an injury as a 'bodily lesion at the organic level, resulting from acute exposure to energy (mechanical, thermal, electrical, chemical or radiant), in amounts that exceed the threshold of physiological tolerance.' We will explore these and other definitions with a view to establishing a more coherent understanding of the ontology of injury and of related phenomena such as lesion, trauma, pain, and so forth. Further topics for possible discussion are listed <u>[[Injury | here]]</u>. For the ontology of pain, see in particular [https://philpapers.org/archive/SMITAO-12.pdf here]<br />
<br />
=='''Draft Agenda: Friday Feb 24'''==<br />
<br />
09:00-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health.<br />
<br />
The ontological representation of occupation presents several challenges. For example, is it your occupation if you have not yet done it? Are you a nuclear engineer before you start your first job doing it? Also, if an occupation is a role (in the BFO sense), then how does it differ from a job role? If I have an occupation of professor, and have had four job roles as professors at different organizations, then that is one occupation numerically but four job roles. How to handle non-employment situations such as "retired" and "homemaker"? What if you concurrently hold two jobs in different "occupations"? I will discuss these issues in the context of recent discussions at the OMRSE monthly meeting and present some options for ontological representation of occupation and related entities.<br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
*'''Justin Reese:''' Unsupervised machine learning to define subtypes of long COVID using the Human Phenotype Ontology. [[Authors | Co-authors]] <br />
<br />
'''Abstract:''' Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterized by a wide range of manifestations that are difficult to analyze computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.<br />
<br />
We present a method for defining subtypes of long COVID by computationally modeling PASC phenotype data and applying unsupervised machine learning. We extracted Long COVID phenotype data expressed as HPO (Human Phenotype Ontology) terms from electronic healthcare records (EHRs), and then used semantic similarity of phenotype data to calculate a matrix of pairwise patient similarity. This matrix was then used to clustered patients using unsupervised machine learning (k means clustering).<br />
<br />
We identified six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centers to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems.<br />
<br />
Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.<br />
<br />
12:30-01:30p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations from industry partners<br />
<br />
TBD<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations <br />
<br />
== '''Participants''' ==<br />
<br />
* Alekseyenko, Alex, Medical University of South Carolina<br />
* Baker, Hamilton, Medical University of South Carolina<br />
* Doole, John, TriNetX<br />
* Heider, Paul, Medical University of South Carolina<br />
* Landgrebe, Jobst, Cognotekt, Cologne<br />
* Hogan, William, co-organizer, University of Florida<br />
* Hutchinson, Tom, University of Pennsylvania, Institute for BioInformatics<br />
* Masci, Anna Maria, National Institute of Environmental Health Sciences (NIEHS)<br />
* Obeid, Jihad, co-organizer, Medical University of South Carolina<br />
* Ohrbach, Richard, University at Buffalo<br />
* Reese, Justin, Lawrence Berkeley National Laboratory<br />
* Scheuermann, Richard, J. Craig Venter Institute<br />
* Simpson, Kit, Medical University of South Carolina<br />
* Skowronek, Matt, TriNetX<br />
* Smith, Barry, co-organizer, University at Buffalo<br />
* Topaloglu, Umit, Center for Biomedical Informatics and Information Technology (CBIIT) National Cancer Institute<br />
* Yonghui Wu, University of Florida</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73403CTS Ontology Workshop 20232023-02-13T18:26:19Z<p>Jihad: /* Participants */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
[[File:Chs23sm.jpg|link=|right|alt=Charleston]]<br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
* Social Determinants of Health<br />
* Mental health<br />
* ChatGPT<br />
<br />
== '''Organizers''' ==<br />
<br />
'''Workshop Co-organizers'''<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
'''CTSOG Co-chairs'''<br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* '''Medical University of South Carolina, the Biomedical Informatics Center''' and:<br />
<br />
[[File:Markiii-logo_160.png|link=https://www.markiiisys.com/|Mark III Systems]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:Nvidia-logo_180.png|link=https://www.nvidia.com/en-us/|NVIDIA]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:TriNetX-Logo-Horizontal-White_180.jpg|link=http://trinetx.com/|TriNetX]]<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Draft Agenda: Thursday Feb 23'''==<br />
<br />
08:30a: Breakfast and Networking<br />
<br />
09:00a: Introduction to the meeting<br />
<br />
09:15-11:00a Working sessions/Presentations<br />
<br />
*'''Barry Smith and Jobst Landgrebe:''' Discussion on AI and Medicine, with a special focus on ChatGPT<br />
<br />
'''Abstract: '''How are we to explain the peculiar tendency of ChatGPT to throw up what are called [https://buffalo.box.com/v/ChatGPT-Hallucinations ‘hallucinations’]? To answer this question we will draw on our book ''[https://buffalo.app.box.com/v/AI-Without-Fear Why Machines Will Never Rule the World]'', whose core thesis is that an artificial intelligence that could equal or exceed human intelligence—sometimes called "artificial general intelligence" (AGI)—is for mathematical reasons impossible. The argument for this thesis rests on the fact that (1.) human intelligence is a capability of a complex dynamic system—the human brain and central nervous system, and (2.) systems of this sort (like all organic systems) cannot be modelled mathematically in a way that would allow the models to operate inside a computer. We survey on this basis the potential of AI in the future of clinical and translational research. <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations<br />
<br />
*'''Discussion and brief presentations. Topics to include:<br />
:The Clinical and Translational Science ontology landscape in 2023<br />
:Opportunities for AI in medicine<br />
:Will AI make ontologies redundant?<br />
:TBD<br />
<br />
12:30-01:30p Networking and Working Lunch <br />
<br />
01:30-3:00p Working sessions/Presentations<br />
<br />
*'''Yonghui Wu:''' A large language model for electronic health records.<br />
<br />
'''Abstract:''' Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
--break--<br />
<br />
3:15-5:00p Working sessions/Presentations<br />
<br />
*'''Richard Ohrbach and Barry Smith:''' Defining 'Injury'<br />
<br />
'''Abstract:''' This session is intended as a contribution to the ontology component of a broader investigation of injury and pain. SNOMED-CT defines an injury as a 'disorder resulting from physical damage to the body'. The WHO defines an injury as a 'bodily lesion at the organic level, resulting from acute exposure to energy (mechanical, thermal, electrical, chemical or radiant), in amounts that exceed the threshold of physiological tolerance.' We will explore these and other definitions with a view to establishing a more coherent understanding of the ontology of injury and of related phenomena such as lesion, trauma, pain, and so forth. Further topics for possible discussion are listed <u>[[Injury | here]]</u>. For the ontology of pain, see in particular [https://philpapers.org/archive/SMITAO-12.pdf here]<br />
<br />
=='''Draft Agenda: Friday Feb 24'''==<br />
<br />
09:00-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health.<br />
<br />
The ontological representation of occupation presents several challenges. For example, is it your occupation if you have not yet done it? Are you a nuclear engineer before you start your first job doing it? Also, if an occupation is a role (in the BFO sense), then how does it differ from a job role? If I have an occupation of professor, and have had four job roles as professors at different organizations, then that is one occupation numerically but four job roles. How to handle non-employment situations such as "retired" and "homemaker"? What if you concurrently hold two jobs in different "occupations"? I will discuss these issues in the context of recent discussions at the OMRSE monthly meeting and present some options for ontological representation of occupation and related entities.<br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
*'''Justin Reese:''' Unsupervised machine learning to define subtypes of long COVID using the Human Phenotype Ontology. [[Authors | Co-authors]] <br />
<br />
'''Abstract:''' Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterized by a wide range of manifestations that are difficult to analyze computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.<br />
<br />
We present a method for defining subtypes of long COVID by computationally modeling PASC phenotype data and applying unsupervised machine learning. We extracted Long COVID phenotype data expressed as HPO (Human Phenotype Ontology) terms from electronic healthcare records (EHRs), and then used semantic similarity of phenotype data to calculate a matrix of pairwise patient similarity. This matrix was then used to clustered patients using unsupervised machine learning (k means clustering).<br />
<br />
We identified six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centers to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems.<br />
<br />
Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.<br />
<br />
12:30-01:30p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations from industry partners<br />
<br />
TBD<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations <br />
<br />
== '''Participants''' ==<br />
<br />
* William Hogan, co-organizer, University of Florida<br />
* Jihad Obeid, co-organizer, Medical University of South Carolina<br />
* Barry Smith, co-organizer, University at Buffalo<br />
<br />
* Hamilton Baker, Medical University of South Carolina<br />
* Heider, Paul, Medical University of South Carolina<br />
* Jobst Landgrebe, Cognotekt, Cologne<br />
* Tom Hutchinson, University of Pennsylvania, Institute for BioInformatics<br />
* Anna Maria Masci, National Institute of Environmental Health Sciences (NIEHS)<br />
* Richard Ohrbach, University at Buffalo<br />
* Justin Reese, Lawrence Berkeley National Laboratory<br />
* Richard Scheuermann, J. Craig Venter Institute<br />
* Matt Skowronek, TriNetX<br />
* Umit Topaloglu, Center for Biomedical Informatics and Information Technology (CBIIT) National Cancer Institute<br />
* Yonghui Wu, University of Florida<br />
* Doole, John, TriNetX<br />
* Simpson, Kit, Medical University of South Carolina<br />
* Alekseyenko, Alex, Medical University of South Carolina</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73392CTS Ontology Workshop 20232023-02-13T16:42:15Z<p>Jihad: /* Participants */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
[[File:Chs23sm.jpg|link=|right|alt=Charleston]]<br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
* Social Determinants of Health<br />
* Mental health<br />
* ChatGPT<br />
<br />
== '''Organizers''' ==<br />
<br />
'''Workshop Co-organizers'''<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
'''CTSOG Co-chairs'''<br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* '''Medical University of South Carolina, the Biomedical Informatics Center''' and:<br />
<br />
[[File:Markiii-logo_160.png|link=https://www.markiiisys.com/|Mark III Systems]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:Nvidia-logo_180.png|link=https://www.nvidia.com/en-us/|NVIDIA]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:TriNetX-Logo-Horizontal-White_180.jpg|link=http://trinetx.com/|TriNetX]]<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Draft Agenda: Thursday Feb 23'''==<br />
<br />
08:30a: Breakfast and Networking<br />
<br />
09:00a: Introduction to the meeting<br />
<br />
09:15-11:00a Working sessions/Presentations<br />
<br />
*'''Barry Smith and Jobst Landgrebe:''' Discussion on AI and Medicine, with a special focus on ChatGPT<br />
<br />
'''Abstract: '''How are we to explain the peculiar tendency of ChatGPT to throw up what are called [https://buffalo.box.com/v/ChatGPT-Hallucinations ‘hallucinations’]? To answer this question we will draw on our book ''[https://buffalo.app.box.com/v/AI-Without-Fear Why Machines Will Never Rule the World]'', whose core thesis is that an artificial intelligence that could equal or exceed human intelligence—sometimes called "artificial general intelligence" (AGI)—is for mathematical reasons impossible. The argument for this thesis rests on the fact that (1.) human intelligence is a capability of a complex dynamic system—the human brain and central nervous system, and (2.) systems of this sort (like all organic systems) cannot be modelled mathematically in a way that would allow the models to operate inside a computer. We survey on this basis the potential of AI in the future of clinical and translational research. <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations<br />
<br />
*'''Discussion and brief presentations. Topics to include:<br />
:The Clinical and Translational Science ontology landscape in 2023<br />
:Opportunities for AI in medicine<br />
:Will AI make ontologies redundant?<br />
:TBD<br />
<br />
12:30-01:30p Networking and Working Lunch <br />
<br />
01:30-3:00p Working sessions/Presentations<br />
<br />
*'''Yonghui Wu:''' A large language model for electronic health records.<br />
<br />
'''Abstract:''' Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
--break--<br />
<br />
3:15-5:00p Working sessions/Presentations<br />
<br />
*'''Richard Ohrbach and Barry Smith:''' Defining 'Injury'<br />
<br />
'''Abstract:''' This session is intended as a contribution to the ontology component of a broader investigation of injury and pain. SNOMED-CT defines an injury as a 'disorder resulting from physical damage to the body'. The WHO defines an injury as a 'bodily lesion at the organic level, resulting from acute exposure to energy (mechanical, thermal, electrical, chemical or radiant), in amounts that exceed the threshold of physiological tolerance.' We will explore these and other definitions with a view to establishing a more coherent understanding of the ontology of injury and of related phenomena such as lesion, trauma, pain, and so forth. Further topics for possible discussion are listed <u>[[Injury | here]]</u>. For the ontology of pain, see in particular [https://philpapers.org/archive/SMITAO-12.pdf here]<br />
<br />
=='''Draft Agenda: Friday Feb 24'''==<br />
<br />
09:00-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health.<br />
<br />
The ontological representation of occupation presents several challenges. For example, is it your occupation if you have not yet done it? Are you a nuclear engineer before you start your first job doing it? Also, if an occupation is a role (in the BFO sense), then how does it differ from a job role? If I have an occupation of professor, and have had four job roles as professors at different organizations, then that is one occupation numerically but four job roles. How to handle non-employment situations such as "retired" and "homemaker"? What if you concurrently hold two jobs in different "occupations"? I will discuss these issues in the context of recent discussions at the OMRSE monthly meeting and present some options for ontological representation of occupation and related entities.<br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
*'''Justin Reese:''' Unsupervised machine learning to define subtypes of long COVID using the Human Phenotype Ontology. [[Authors | Co-authors]] <br />
<br />
'''Abstract:''' Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterized by a wide range of manifestations that are difficult to analyze computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.<br />
<br />
We present a method for defining subtypes of long COVID by computationally modeling PASC phenotype data and applying unsupervised machine learning. We extracted Long COVID phenotype data expressed as HPO (Human Phenotype Ontology) terms from electronic healthcare records (EHRs), and then used semantic similarity of phenotype data to calculate a matrix of pairwise patient similarity. This matrix was then used to clustered patients using unsupervised machine learning (k means clustering).<br />
<br />
We identified six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centers to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems.<br />
<br />
Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.<br />
<br />
12:30-01:30p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations from industry partners<br />
<br />
TBD<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations <br />
<br />
== '''Participants''' ==<br />
<br />
* William Hogan, co-organizer, University of Florida<br />
* Jihad Obeid, co-organizer, Medical University of South Carolina<br />
* Barry Smith, co-organizer, University at Buffalo<br />
<br />
* Hamilton Baker, Medical University of South Carolina<br />
* Heider, Paul, Medical University of South Carolina<br />
* Jobst Landgrebe, Cognotekt, Cologne<br />
* Tom Hutchinson, University of Pennsylvania, Institute for BioInformatics<br />
* Anna Maria Masci, National Institute of Environmental Health Sciences (NIEHS)<br />
* Richard Ohrbach, University at Buffalo<br />
* Justin Reese, Lawrence Berkeley National Laboratory<br />
* Richard Scheuermann, J. Craig Venter Institute<br />
* Matt Skowronek, TriNetX<br />
* Umit Topaloglu, Center for Biomedical Informatics and Information Technology (CBIIT) National Cancer Institute<br />
* Yonghui Wu, University of Florida<br />
* Doole, John, TriNetX<br />
* Simpson, Kit, Medical University of South Carolina</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73391CTS Ontology Workshop 20232023-02-13T16:14:23Z<p>Jihad: /* Participants */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
[[File:Chs23sm.jpg|link=|right|alt=Charleston]]<br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
* Social Determinants of Health<br />
* Mental health<br />
* ChatGPT<br />
<br />
== '''Organizers''' ==<br />
<br />
'''Workshop Co-organizers'''<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
'''CTSOG Co-chairs'''<br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* '''Medical University of South Carolina, the Biomedical Informatics Center''' and:<br />
<br />
[[File:Markiii-logo_160.png|link=https://www.markiiisys.com/|Mark III Systems]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:Nvidia-logo_180.png|link=https://www.nvidia.com/en-us/|NVIDIA]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:TriNetX-Logo-Horizontal-White_180.jpg|link=http://trinetx.com/|TriNetX]]<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Draft Agenda: Thursday Feb 23'''==<br />
<br />
08:30a: Breakfast and Networking<br />
<br />
09:00a: Introduction to the meeting<br />
<br />
09:15-11:00a Working sessions/Presentations<br />
<br />
*'''Barry Smith and Jobst Landgrebe:''' Discussion on AI and Medicine, with a special focus on ChatGPT<br />
<br />
'''Abstract: '''How are we to explain the peculiar tendency of ChatGPT to throw up what are called [https://buffalo.box.com/v/ChatGPT-Hallucinations ‘hallucinations’]? To answer this question we will draw on our book ''[https://buffalo.app.box.com/v/AI-Without-Fear Why Machines Will Never Rule the World]'', whose core thesis is that an artificial intelligence that could equal or exceed human intelligence—sometimes called "artificial general intelligence" (AGI)—is for mathematical reasons impossible. The argument for this thesis rests on the fact that (1.) human intelligence is a capability of a complex dynamic system—the human brain and central nervous system, and (2.) systems of this sort (like all organic systems) cannot be modelled mathematically in a way that would allow the models to operate inside a computer. We survey on this basis the potential of AI in the future of clinical and translational research. <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations<br />
<br />
*'''Discussion and brief presentations. Topics to include:<br />
:The Clinical and Translational Science ontology landscape in 2023<br />
:Opportunities for AI in medicine<br />
:Will AI make ontologies redundant?<br />
:TBD<br />
<br />
12:30-01:30p Networking and Working Lunch <br />
<br />
01:30-3:00p Working sessions/Presentations<br />
<br />
*'''Yonghui Wu:''' A large language model for electronic health records.<br />
<br />
'''Abstract:''' Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
--break--<br />
<br />
3:15-5:00p Working sessions/Presentations<br />
<br />
*'''Richard Ohrbach and Barry Smith:''' Defining 'Injury'<br />
<br />
'''Abstract:''' This session is intended as a contribution to the ontology component of a broader investigation of injury and pain. SNOMED-CT defines an injury as a 'disorder resulting from physical damage to the body'. The WHO defines an injury as a 'bodily lesion at the organic level, resulting from acute exposure to energy (mechanical, thermal, electrical, chemical or radiant), in amounts that exceed the threshold of physiological tolerance.' We will explore these and other definitions with a view to establishing a more coherent understanding of the ontology of injury and of related phenomena such as lesion, trauma, pain, and so forth. Further topics for possible discussion are listed <u>[[Injury | here]]</u>. For the ontology of pain, see in particular [https://philpapers.org/archive/SMITAO-12.pdf here]<br />
<br />
=='''Draft Agenda: Friday Feb 24'''==<br />
<br />
09:00-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health.<br />
<br />
The ontological representation of occupation presents several challenges. For example, is it your occupation if you have not yet done it? Are you a nuclear engineer before you start your first job doing it? Also, if an occupation is a role (in the BFO sense), then how does it differ from a job role? If I have an occupation of professor, and have had four job roles as professors at different organizations, then that is one occupation numerically but four job roles. How to handle non-employment situations such as "retired" and "homemaker"? What if you concurrently hold two jobs in different "occupations"? I will discuss these issues in the context of recent discussions at the OMRSE monthly meeting and present some options for ontological representation of occupation and related entities.<br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
*'''Justin Reese:''' Unsupervised machine learning to define subtypes of long COVID using the Human Phenotype Ontology. [[Authors | Co-authors]] <br />
<br />
'''Abstract:''' Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterized by a wide range of manifestations that are difficult to analyze computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.<br />
<br />
We present a method for defining subtypes of long COVID by computationally modeling PASC phenotype data and applying unsupervised machine learning. We extracted Long COVID phenotype data expressed as HPO (Human Phenotype Ontology) terms from electronic healthcare records (EHRs), and then used semantic similarity of phenotype data to calculate a matrix of pairwise patient similarity. This matrix was then used to clustered patients using unsupervised machine learning (k means clustering).<br />
<br />
We identified six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centers to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems.<br />
<br />
Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.<br />
<br />
12:30-01:30p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations from industry partners<br />
<br />
TBD<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations <br />
<br />
== '''Participants''' ==<br />
<br />
* William Hogan, co-organizer, University of Florida<br />
* Jihad Obeid, co-organizer, Medical University of South Carolina<br />
* Barry Smith, co-organizer, University at Buffalo<br />
<br />
* Hamilton Baker, Medical University of South Carolina<br />
* Heider, Paul, Medical University of South Carolina<br />
* Jobst Landgrebe, Cognotekt, Cologne<br />
* Tom Hutchinson, University of Pennsylvania, Institute for BioInformatics<br />
* Anna Maria Masci, National Institute of Environmental Health Sciences (NIEHS)<br />
* Richard Ohrbach, University at Buffalo<br />
* Justin Reese, Lawrence Berkeley National Laboratory<br />
* Richard Scheuermann, J. Craig Venter Institute<br />
* Matt Skowronek, TriNetX<br />
* Umit Topaloglu, Center for Biomedical Informatics and Information Technology (CBIIT) National Cancer Institute<br />
* Yonghui Wu, University of Florida<br />
* Doole, John, TriNetX</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73390CTS Ontology Workshop 20232023-02-09T18:16:03Z<p>Jihad: /* Participants */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
[[File:Chs23sm.jpg|link=|right|alt=Charleston]]<br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
* Social Determinants of Health<br />
* Mental health<br />
* ChatGPT<br />
<br />
== '''Organizers''' ==<br />
<br />
'''Workshop Co-organizers'''<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
'''CTSOG Co-chairs'''<br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* '''Medical University of South Carolina, the Biomedical Informatics Center''' and:<br />
<br />
[[File:Markiii-logo_160.png|link=https://www.markiiisys.com/|Mark III Systems]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:Nvidia-logo_180.png|link=https://www.nvidia.com/en-us/|NVIDIA]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:TriNetX-Logo-Horizontal-White_180.jpg|link=http://trinetx.com/|TriNetX]]<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Draft Agenda: Thursday Feb 23'''==<br />
<br />
08:30a: Breakfast and Networking<br />
<br />
09:00a: Introduction to the meeting<br />
<br />
09:15-11:00a Working sessions/Presentations<br />
<br />
*'''Barry Smith and Jobst Landgrebe:''' Discussion on AI and Medicine, with a special focus on ChatGPT<br />
<br />
'''Abstract: '''How are we to explain the peculiar tendency of ChatGPT to throw up what are called [https://buffalo.box.com/v/ChatGPT-Hallucinations ‘hallucinations’]? To answer this question we will draw on our book ''[https://buffalo.app.box.com/v/AI-Without-Fear Why Machines Will Never Rule the World]'', whose core thesis is that an artificial intelligence that could equal or exceed human intelligence—sometimes called "artificial general intelligence" (AGI)—is for mathematical reasons impossible. The argument for this thesis rests on the fact that (1.) human intelligence is a capability of a complex dynamic system—the human brain and central nervous system, and (2.) systems of this sort (like all organic systems) cannot be modelled mathematically in a way that would allow the models to operate inside a computer. We survey on this basis the potential of AI in the future of clinical and translational research. <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations<br />
<br />
*'''Discussion and brief presentations. Topics to include:<br />
:The Clinical and Translational Science ontology landscape in 2023<br />
:Opportunities for AI in medicine<br />
:Will AI make ontologies redundant?<br />
:TBD<br />
<br />
12:30-01:30p Networking and Working Lunch <br />
<br />
01:30-3:00p Working sessions/Presentations<br />
<br />
*'''Yonghui Wu:''' A large language model for electronic health records.<br />
<br />
'''Abstract:''' Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
--break--<br />
<br />
3:15-5:00p Working sessions/Presentations<br />
<br />
*'''Richard Ohrbach and Barry Smith:''' Defining 'Injury'<br />
<br />
'''Abstract:''' This session is intended as a contribution to the ontology component of a broader investigation of injury and pain. SNOMED-CT defines an injury as a 'disorder resulting from physical damage to the body'. The WHO defines an injury as a 'bodily lesion at the organic level, resulting from acute exposure to energy (mechanical, thermal, electrical, chemical or radiant), in amounts that exceed the threshold of physiological tolerance.' We will explore these and other definitions with a view to establishing a more coherent understanding of the ontology of injury and of related phenomena such as lesion, trauma, pain, and so forth. Further topics for possible discussion are listed <u>[[Injury | here]]</u>. For the ontology of pain, see in particular [https://philpapers.org/archive/SMITAO-12.pdf here]<br />
<br />
=='''Draft Agenda: Friday Feb 24'''==<br />
<br />
09:00-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health.<br />
<br />
The ontological representation of occupation presents several challenges. For example, is it your occupation if you have not yet done it? Are you a nuclear engineer before you start your first job doing it? Also, if an occupation is a role (in the BFO sense), then how does it differ from a job role? If I have an occupation of professor, and have had four job roles as professors at different organizations, then that is one occupation numerically but four job roles. How to handle non-employment situations such as "retired" and "homemaker"? What if you concurrently hold two jobs in different "occupations"? I will discuss these issues in the context of recent discussions at the OMRSE monthly meeting and present some options for ontological representation of occupation and related entities.<br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
*'''Justin Reese:''' Unsupervised machine learning to define subtypes of long COVID using the Human Phenotype Ontology. [[Authors | Co-authors]] <br />
<br />
'''Abstract:''' Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterized by a wide range of manifestations that are difficult to analyze computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.<br />
<br />
We present a method for defining subtypes of long COVID by computationally modeling PASC phenotype data and applying unsupervised machine learning. We extracted Long COVID phenotype data expressed as HPO (Human Phenotype Ontology) terms from electronic healthcare records (EHRs), and then used semantic similarity of phenotype data to calculate a matrix of pairwise patient similarity. This matrix was then used to clustered patients using unsupervised machine learning (k means clustering).<br />
<br />
We identified six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centers to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems.<br />
<br />
Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.<br />
<br />
12:30-01:30p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations from industry partners<br />
<br />
TBD<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations <br />
<br />
== '''Participants''' ==<br />
<br />
* William Hogan, co-organizer, University of Florida<br />
* Jihad Obeid, co-organizer, Medical University of South Carolina<br />
* Barry Smith, co-organizer, University at Buffalo<br />
<br />
* Hamilton Baker, Medical University of South Carolina<br />
* Heider, Paul, Medical University of South Carolina<br />
* Jobst Landgrebe, Cognotekt, Cologne<br />
* Tom Hutchinson, University of Pennsylvania, Institute for BioInformatics<br />
* Anna Maria Masci, National Institute of Environmental Health Sciences (NIEHS)<br />
* Richard Ohrbach, University at Buffalo<br />
* Justin Reese, Lawrence Berkeley National Laboratory<br />
* Richard Scheuermann, J. Craig Venter Institute<br />
* Matt Skowronek, TriNetX<br />
* Umit Topaloglu, Center for Biomedical Informatics and Information Technology (CBIIT) National Cancer Institute<br />
* Yonghui Wu, University of Florida</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73370CTS Ontology Workshop 20232023-01-31T22:38:45Z<p>Jihad: /* Participants */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
[[File:Chs23sm.jpg|link=|right|alt=Charleston]]<br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
* Social Determinants of Health<br />
* Mental health<br />
* ChatGPT<br />
<br />
== '''Organizers''' ==<br />
<br />
'''Workshop Co-organizers'''<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
'''CTSOG Co-chairs'''<br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* '''Medical University of South Carolina, the Biomedical Informatics Center''' and:<br />
<br />
[[File:Markiii-logo_160.png|link=https://www.markiiisys.com/|Mark III Systems]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:Nvidia-logo_180.png|link=https://www.nvidia.com/en-us/|NVIDIA]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:TriNetX-Logo-Horizontal-White_180.jpg|link=http://trinetx.com/|TriNetX]]<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Agenda: Thursday Feb 23'''==<br />
<br />
08:30a: Breakfast and Networking<br />
<br />
09:00: Introduction to the meeting<br />
<br />
09:15 <br />
<br />
*'''Barry Smith and Jobst Landgrebe:''' Discussion on AI and Medicine, with a special focus on ChatGPT<br />
<br />
'''Abstract: '''How are we to explain the peculiar tendency of ChatGPT to throw up what are called [https://buffalo.box.com/v/ChatGPT-Hallucinations ‘hallucinations’]? To answer this question we will draw on our book ''[https://buffalo.app.box.com/v/AI-Without-Fear Why Machines Will Never Rule the World]'', whose core thesis is that an artificial intelligence that could equal or exceed human intelligence—sometimes called "artificial general intelligence" (AGI)—is for mathematical reasons impossible. The argument for this thesis rests on the fact that (1.) human intelligence is a capability of a complex dynamic system—the human brain and central nervous system, and (2.) systems of this sort (like all organic systems) cannot be modelled mathematically in a way that would allow the models to operate inside a computer. We survey on this basis the potential of AI in the future of clinical and translational research. <br />
<br />
--break--<br />
<br />
11:00<br />
<br />
*'''Discussion and brief presentations on:<br />
:The Clinical and Translational Science ontology landscape in 2023<br />
:Opportunities for AI in medicine<br />
:Will AI make ontologies redundant?<br />
:TBD<br />
<br />
12:30-01:30p Networking and Working Lunch <br />
<br />
01:30-3:00p Working sessions/Presentations<br />
<br />
*'''Yonghui Wu:''' A large language model for electronic health records.<br />
<br />
'''Abstract:''' Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
--break--<br />
<br />
3:15-5:00 <br />
<br />
*'''Richard Ohrbach and Barry Smith:''' Defining 'Injury'<br />
<br />
'''Abstract:''' This session is intended as a contribution to the ontology component of a project for investigating injury and pain. It responds to the NIH [https://www.nidcr.nih.gov/grants-funding/funding-priorities/future-research-initiatives/tmd-collaborative-improving-patientcentered-translational-research-tmd-impact TMD IMPACT Collaborative for IMproving PAtient-Centered Translational Research]. SNOMED-CT defines an injury as a 'disorder resulting from physical damage to the body'. The WHO defines an injury as a 'bodily lesion at the organic level, resulting from acute exposure to energy (mechanical, thermal, electrical, chemical or radiant), in amounts that exceed the threshold of physiological tolerance.' We will explore these and other definitions with a view to establishing a more coherent understanding of the ontology of injury and of related phenomena such as lesion, trauma, pain, and so forth. Further topics are listed [[Injury | here]].<br />
<br />
=='''Agenda: Friday Feb 24'''==<br />
<br />
09:00-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health.<br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
*'''Justin Reese:''' Unsupervised machine learning to define subtypes of long COVID using the Human Phenotype Ontology. [[Authors | Co-authors]] <br />
<br />
'''Abstract:''' Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterized by a wide range of manifestations that are difficult to analyze computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.<br />
<br />
We present a method for defining subtypes of long COVID by computationally modeling PASC phenotype data and applying unsupervised machine learning. We extracted Long COVID phenotype data expressed as HPO (Human Phenotype Ontology) terms from electronic healthcare records (EHRs), and then used semantic similarity of phenotype data to calculate a matrix of pairwise patient similarity. This matrix was then used to clustered patients using unsupervised machine learning (k means clustering).<br />
<br />
We identified six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centers to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems.<br />
<br />
Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.<br />
<br />
12:30-01:30p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations of industry partners<br />
<br />
== '''Participants''' ==<br />
<br />
* William Hogan, co-organizer, University of Florida<br />
* Jihad Obeid, co-organizer, Medical University of South Carolina<br />
* Barry Smith, co-organizer, University at Buffalo<br />
<br />
* Hamilton Baker, Medical University of South Carolina<br />
* Jobst Landgrebe, Cognotekt, Cologne<br />
* Tom Hutchinson, University of Pennsylvania, Institute for BioInformatics<br />
* Anna Maria Masci, National Institute of Environmental Health Sciences (NIEHS)<br />
* Richard Ohrbach, University at Buffalo<br />
* Justin Reese, Lawrence Berkeley National Laboratory<br />
* Richard Scheuermann, J. Craig Venter Institute<br />
* Matt Skowronek, TriNetX<br />
* Umit Topaloglu, Center for Biomedical Informatics and Information Technology (CBIIT) National Cancer Institute<br />
* Yonghui Wu, University of Florida</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73369CTS Ontology Workshop 20232023-01-31T22:38:22Z<p>Jihad: /* Participants */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
[[File:Chs23sm.jpg|link=|right|alt=Charleston]]<br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
* Social Determinants of Health<br />
* Mental health<br />
* ChatGPT<br />
<br />
== '''Organizers''' ==<br />
<br />
'''Workshop Co-organizers'''<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
'''CTSOG Co-chairs'''<br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* '''Medical University of South Carolina, the Biomedical Informatics Center''' and:<br />
<br />
[[File:Markiii-logo_160.png|link=https://www.markiiisys.com/|Mark III Systems]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:Nvidia-logo_180.png|link=https://www.nvidia.com/en-us/|NVIDIA]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:TriNetX-Logo-Horizontal-White_180.jpg|link=http://trinetx.com/|TriNetX]]<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Agenda: Thursday Feb 23'''==<br />
<br />
08:30a: Breakfast and Networking<br />
<br />
09:00: Introduction to the meeting<br />
<br />
09:15 <br />
<br />
*'''Barry Smith and Jobst Landgrebe:''' Discussion on AI and Medicine, with a special focus on ChatGPT<br />
<br />
'''Abstract: '''How are we to explain the peculiar tendency of ChatGPT to throw up what are called [https://buffalo.box.com/v/ChatGPT-Hallucinations ‘hallucinations’]? To answer this question we will draw on our book ''[https://buffalo.app.box.com/v/AI-Without-Fear Why Machines Will Never Rule the World]'', whose core thesis is that an artificial intelligence that could equal or exceed human intelligence—sometimes called "artificial general intelligence" (AGI)—is for mathematical reasons impossible. The argument for this thesis rests on the fact that (1.) human intelligence is a capability of a complex dynamic system—the human brain and central nervous system, and (2.) systems of this sort (like all organic systems) cannot be modelled mathematically in a way that would allow the models to operate inside a computer. We survey on this basis the potential of AI in the future of clinical and translational research. <br />
<br />
--break--<br />
<br />
11:00<br />
<br />
*'''Discussion and brief presentations on:<br />
:The Clinical and Translational Science ontology landscape in 2023<br />
:Opportunities for AI in medicine<br />
:Will AI make ontologies redundant?<br />
:TBD<br />
<br />
12:30-01:30p Networking and Working Lunch <br />
<br />
01:30-3:00p Working sessions/Presentations<br />
<br />
*'''Yonghui Wu:''' A large language model for electronic health records.<br />
<br />
'''Abstract:''' Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
--break--<br />
<br />
3:15-5:00 <br />
<br />
*'''Richard Ohrbach and Barry Smith:''' Defining 'Injury'<br />
<br />
'''Abstract:''' This session is intended as a contribution to the ontology component of a project for investigating injury and pain. It responds to the NIH [https://www.nidcr.nih.gov/grants-funding/funding-priorities/future-research-initiatives/tmd-collaborative-improving-patientcentered-translational-research-tmd-impact TMD IMPACT Collaborative for IMproving PAtient-Centered Translational Research]. SNOMED-CT defines an injury as a 'disorder resulting from physical damage to the body'. The WHO defines an injury as a 'bodily lesion at the organic level, resulting from acute exposure to energy (mechanical, thermal, electrical, chemical or radiant), in amounts that exceed the threshold of physiological tolerance.' We will explore these and other definitions with a view to establishing a more coherent understanding of the ontology of injury and of related phenomena such as lesion, trauma, pain, and so forth. Further topics are listed [[Injury | here]].<br />
<br />
=='''Agenda: Friday Feb 24'''==<br />
<br />
09:00-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health.<br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
*'''Justin Reese:''' Unsupervised machine learning to define subtypes of long COVID using the Human Phenotype Ontology. [[Authors | Co-authors]] <br />
<br />
'''Abstract:''' Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterized by a wide range of manifestations that are difficult to analyze computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.<br />
<br />
We present a method for defining subtypes of long COVID by computationally modeling PASC phenotype data and applying unsupervised machine learning. We extracted Long COVID phenotype data expressed as HPO (Human Phenotype Ontology) terms from electronic healthcare records (EHRs), and then used semantic similarity of phenotype data to calculate a matrix of pairwise patient similarity. This matrix was then used to clustered patients using unsupervised machine learning (k means clustering).<br />
<br />
We identified six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centers to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems.<br />
<br />
Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.<br />
<br />
12:30-01:30p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations of industry partners<br />
<br />
== '''Participants''' ==<br />
<br />
* William Hogan, co-organizer, University of Florida<br />
* Jihad Obeid, co-organizer, Medical University of South Carolina<br />
* Barry Smith, co-organizer, University at Buffalo<br />
<br />
* Hamilton Baker, Medical University of South Carolina<br />
* Jobst Landgrebe, Cognotekt, Cologne<br />
* Tom Hutchinson, University of Pennsylvania, Institute for BioInformatics<br />
* Anna Maria Masci, National Institute of Environmental Health Sciences (NIEHS)<br />
* Richard Ohrbach, University at Buffalo<br />
* Justin Reese, Lawrence Berkeley National Laboratory<br />
* Richard Scheuermann, J. Craig Venter Institute<br />
* Matt Skowronek, TriNetX<br />
* Umit Topaloglu, Center for Biomedical Informatics and Information Technology (CBIIT) National Cancer Institute (NCI)<br />
* Yonghui Wu, University of Florida</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73368CTS Ontology Workshop 20232023-01-31T22:37:35Z<p>Jihad: /* Participants */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
[[File:Chs23sm.jpg|link=|right|alt=Charleston]]<br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
* Social Determinants of Health<br />
* Mental health<br />
* ChatGPT<br />
<br />
== '''Organizers''' ==<br />
<br />
'''Workshop Co-organizers'''<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
'''CTSOG Co-chairs'''<br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* '''Medical University of South Carolina, the Biomedical Informatics Center''' and:<br />
<br />
[[File:Markiii-logo_160.png|link=https://www.markiiisys.com/|Mark III Systems]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:Nvidia-logo_180.png|link=https://www.nvidia.com/en-us/|NVIDIA]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:TriNetX-Logo-Horizontal-White_180.jpg|link=http://trinetx.com/|TriNetX]]<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Agenda: Thursday Feb 23'''==<br />
<br />
08:30a: Breakfast and Networking<br />
<br />
09:00: Introduction to the meeting<br />
<br />
09:15 <br />
<br />
*'''Barry Smith and Jobst Landgrebe:''' Discussion on AI and Medicine, with a special focus on ChatGPT<br />
<br />
'''Abstract: '''How are we to explain the peculiar tendency of ChatGPT to throw up what are called [https://buffalo.box.com/v/ChatGPT-Hallucinations ‘hallucinations’]? To answer this question we will draw on our book ''[https://buffalo.app.box.com/v/AI-Without-Fear Why Machines Will Never Rule the World]'', whose core thesis is that an artificial intelligence that could equal or exceed human intelligence—sometimes called "artificial general intelligence" (AGI)—is for mathematical reasons impossible. The argument for this thesis rests on the fact that (1.) human intelligence is a capability of a complex dynamic system—the human brain and central nervous system, and (2.) systems of this sort (like all organic systems) cannot be modelled mathematically in a way that would allow the models to operate inside a computer. We survey on this basis the potential of AI in the future of clinical and translational research. <br />
<br />
--break--<br />
<br />
11:00<br />
<br />
*'''Discussion and brief presentations on:<br />
:The Clinical and Translational Science ontology landscape in 2023<br />
:Opportunities for AI in medicine<br />
:Will AI make ontologies redundant?<br />
:TBD<br />
<br />
12:30-01:30p Networking and Working Lunch <br />
<br />
01:30-3:00p Working sessions/Presentations<br />
<br />
*'''Yonghui Wu:''' A large language model for electronic health records.<br />
<br />
'''Abstract:''' Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
--break--<br />
<br />
3:15-5:00 <br />
<br />
*'''Richard Ohrbach and Barry Smith:''' Defining 'Injury'<br />
<br />
'''Abstract:''' This session is intended as a contribution to the ontology component of a project for investigating injury and pain. It responds to the NIH [https://www.nidcr.nih.gov/grants-funding/funding-priorities/future-research-initiatives/tmd-collaborative-improving-patientcentered-translational-research-tmd-impact TMD IMPACT Collaborative for IMproving PAtient-Centered Translational Research]. SNOMED-CT defines an injury as a 'disorder resulting from physical damage to the body'. The WHO defines an injury as a 'bodily lesion at the organic level, resulting from acute exposure to energy (mechanical, thermal, electrical, chemical or radiant), in amounts that exceed the threshold of physiological tolerance.' We will explore these and other definitions with a view to establishing a more coherent understanding of the ontology of injury and of related phenomena such as lesion, trauma, pain, and so forth. Further topics are listed [[Injury | here]].<br />
<br />
=='''Agenda: Friday Feb 24'''==<br />
<br />
09:00-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health.<br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
*'''Justin Reese:''' Unsupervised machine learning to define subtypes of long COVID using the Human Phenotype Ontology. [[Authors | Co-authors]] <br />
<br />
'''Abstract:''' Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterized by a wide range of manifestations that are difficult to analyze computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.<br />
<br />
We present a method for defining subtypes of long COVID by computationally modeling PASC phenotype data and applying unsupervised machine learning. We extracted Long COVID phenotype data expressed as HPO (Human Phenotype Ontology) terms from electronic healthcare records (EHRs), and then used semantic similarity of phenotype data to calculate a matrix of pairwise patient similarity. This matrix was then used to clustered patients using unsupervised machine learning (k means clustering).<br />
<br />
We identified six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centers to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems.<br />
<br />
Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.<br />
<br />
12:30-01:30p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations of industry partners<br />
<br />
== '''Participants''' ==<br />
<br />
* William Hogan, co-organizer, University of Florida<br />
* Jihad Obeid, co-organizer, Medical University of South Carolina<br />
* Barry Smith, co-organizer, University at Buffalo<br />
<br />
* Hamilton Baker, Medical University of South Carolina<br />
* Jobst Landgrebe, Cognotekt, Cologne<br />
* Tom Hutchinson, University of Pennsylvania, Institute for BioInformatics<br />
* Anna Maria Masci, National Institute of Environmental Health Sciences (NIEHS)<br />
* Richard Ohrbach, University at Buffalo<br />
* Justin Reese, Lawrence Berkeley National Laboratory<br />
* Richard Scheuermann, J. Craig Venter Institute<br />
* Matt Skowronek, TriNetX<br />
* Yonghui Wu, University of Florida</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73367CTS Ontology Workshop 20232023-01-31T22:36:41Z<p>Jihad: /* Participants */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
[[File:Chs23sm.jpg|link=|right|alt=Charleston]]<br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
* Social Determinants of Health<br />
* Mental health<br />
* ChatGPT<br />
<br />
== '''Organizers''' ==<br />
<br />
'''Workshop Co-organizers'''<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
'''CTSOG Co-chairs'''<br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* '''Medical University of South Carolina, the Biomedical Informatics Center''' and:<br />
<br />
[[File:Markiii-logo_160.png|link=https://www.markiiisys.com/|Mark III Systems]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:Nvidia-logo_180.png|link=https://www.nvidia.com/en-us/|NVIDIA]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:TriNetX-Logo-Horizontal-White_180.jpg|link=http://trinetx.com/|TriNetX]]<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Agenda: Thursday Feb 23'''==<br />
<br />
08:30a: Breakfast and Networking<br />
<br />
09:00: Introduction to the meeting<br />
<br />
09:15 <br />
<br />
*'''Barry Smith and Jobst Landgrebe:''' Discussion on AI and Medicine, with a special focus on ChatGPT<br />
<br />
'''Abstract: '''How are we to explain the peculiar tendency of ChatGPT to throw up what are called [https://buffalo.box.com/v/ChatGPT-Hallucinations ‘hallucinations’]? To answer this question we will draw on our book ''[https://buffalo.app.box.com/v/AI-Without-Fear Why Machines Will Never Rule the World]'', whose core thesis is that an artificial intelligence that could equal or exceed human intelligence—sometimes called "artificial general intelligence" (AGI)—is for mathematical reasons impossible. The argument for this thesis rests on the fact that (1.) human intelligence is a capability of a complex dynamic system—the human brain and central nervous system, and (2.) systems of this sort (like all organic systems) cannot be modelled mathematically in a way that would allow the models to operate inside a computer. We survey on this basis the potential of AI in the future of clinical and translational research. <br />
<br />
--break--<br />
<br />
11:00<br />
<br />
*'''Discussion and brief presentations on:<br />
:The Clinical and Translational Science ontology landscape in 2023<br />
:Opportunities for AI in medicine<br />
:Will AI make ontologies redundant?<br />
:TBD<br />
<br />
12:30-01:30p Networking and Working Lunch <br />
<br />
01:30-3:00p Working sessions/Presentations<br />
<br />
*'''Yonghui Wu:''' A large language model for electronic health records.<br />
<br />
'''Abstract:''' Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
--break--<br />
<br />
3:15-5:00 <br />
<br />
*'''Richard Ohrbach and Barry Smith:''' Defining 'Injury'<br />
<br />
'''Abstract:''' This session is intended as a contribution to the ontology component of a project for investigating injury and pain. It responds to the NIH [https://www.nidcr.nih.gov/grants-funding/funding-priorities/future-research-initiatives/tmd-collaborative-improving-patientcentered-translational-research-tmd-impact TMD IMPACT Collaborative for IMproving PAtient-Centered Translational Research]. SNOMED-CT defines an injury as a 'disorder resulting from physical damage to the body'. The WHO defines an injury as a 'bodily lesion at the organic level, resulting from acute exposure to energy (mechanical, thermal, electrical, chemical or radiant), in amounts that exceed the threshold of physiological tolerance.' We will explore these and other definitions with a view to establishing a more coherent understanding of the ontology of injury and of related phenomena such as lesion, trauma, pain, and so forth. Further topics are listed [[Injury | here]].<br />
<br />
=='''Agenda: Friday Feb 24'''==<br />
<br />
09:00-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health.<br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
*'''Justin Reese:''' Unsupervised machine learning to define subtypes of long COVID using the Human Phenotype Ontology. [[Authors | Co-authors]] <br />
<br />
'''Abstract:''' Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterized by a wide range of manifestations that are difficult to analyze computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.<br />
<br />
We present a method for defining subtypes of long COVID by computationally modeling PASC phenotype data and applying unsupervised machine learning. We extracted Long COVID phenotype data expressed as HPO (Human Phenotype Ontology) terms from electronic healthcare records (EHRs), and then used semantic similarity of phenotype data to calculate a matrix of pairwise patient similarity. This matrix was then used to clustered patients using unsupervised machine learning (k means clustering).<br />
<br />
We identified six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centers to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems.<br />
<br />
Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.<br />
<br />
12:30-01:30p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations of industry partners<br />
<br />
== '''Participants''' ==<br />
<br />
* William Hogan, co-organizer, University of Florida<br />
* Jihad Obeid, co-organizer, Medical University of South Carolina<br />
* Barry Smith, co-organizer, University at Buffalo<br />
<br />
<br />
* Hamilton Baker, Medical University of South Carolina<br />
* Jobst Landgrebe, Cognotekt, Cologne<br />
* Tom Hutchinson, University of Pennsylvania, Institute for BioInformatics<br />
* Anna Maria Masci, National Institute of Environmental Health Sciences (NIEHS)<br />
* Richard Ohrbach, University at Buffalo<br />
* Justin Reese, Lawrence Berkeley National Laboratory<br />
* Richard Scheuermann, J. Craig Venter Institute<br />
* Matt Skowronek, TriNetX<br />
* Yonghui Wu, University of Florida</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73366CTS Ontology Workshop 20232023-01-31T22:35:58Z<p>Jihad: /* Participants */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
[[File:Chs23sm.jpg|link=|right|alt=Charleston]]<br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
* Social Determinants of Health<br />
* Mental health<br />
* ChatGPT<br />
<br />
== '''Organizers''' ==<br />
<br />
'''Workshop Co-organizers'''<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
'''CTSOG Co-chairs'''<br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* '''Medical University of South Carolina, the Biomedical Informatics Center''' and:<br />
<br />
[[File:Markiii-logo_160.png|link=https://www.markiiisys.com/|Mark III Systems]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:Nvidia-logo_180.png|link=https://www.nvidia.com/en-us/|NVIDIA]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:TriNetX-Logo-Horizontal-White_180.jpg|link=http://trinetx.com/|TriNetX]]<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Agenda: Thursday Feb 23'''==<br />
<br />
08:30a: Breakfast and Networking<br />
<br />
09:00: Introduction to the meeting<br />
<br />
09:15 <br />
<br />
*'''Barry Smith and Jobst Landgrebe:''' Discussion on AI and Medicine, with a special focus on ChatGPT<br />
<br />
'''Abstract: '''How are we to explain the peculiar tendency of ChatGPT to throw up what are called [https://buffalo.box.com/v/ChatGPT-Hallucinations ‘hallucinations’]? To answer this question we will draw on our book ''[https://buffalo.app.box.com/v/AI-Without-Fear Why Machines Will Never Rule the World]'', whose core thesis is that an artificial intelligence that could equal or exceed human intelligence—sometimes called "artificial general intelligence" (AGI)—is for mathematical reasons impossible. The argument for this thesis rests on the fact that (1.) human intelligence is a capability of a complex dynamic system—the human brain and central nervous system, and (2.) systems of this sort (like all organic systems) cannot be modelled mathematically in a way that would allow the models to operate inside a computer. We survey on this basis the potential of AI in the future of clinical and translational research. <br />
<br />
--break--<br />
<br />
11:00<br />
<br />
*'''Discussion and brief presentations on:<br />
:The Clinical and Translational Science ontology landscape in 2023<br />
:Opportunities for AI in medicine<br />
:Will AI make ontologies redundant?<br />
:TBD<br />
<br />
12:30-01:30p Networking and Working Lunch <br />
<br />
01:30-3:00p Working sessions/Presentations<br />
<br />
*'''Yonghui Wu:''' A large language model for electronic health records.<br />
<br />
'''Abstract:''' Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
--break--<br />
<br />
3:15-5:00 <br />
<br />
*'''Richard Ohrbach and Barry Smith:''' Defining 'Injury'<br />
<br />
'''Abstract:''' This session is intended as a contribution to the ontology component of a project for investigating injury and pain. It responds to the NIH [https://www.nidcr.nih.gov/grants-funding/funding-priorities/future-research-initiatives/tmd-collaborative-improving-patientcentered-translational-research-tmd-impact TMD IMPACT Collaborative for IMproving PAtient-Centered Translational Research]. SNOMED-CT defines an injury as a 'disorder resulting from physical damage to the body'. The WHO defines an injury as a 'bodily lesion at the organic level, resulting from acute exposure to energy (mechanical, thermal, electrical, chemical or radiant), in amounts that exceed the threshold of physiological tolerance.' We will explore these and other definitions with a view to establishing a more coherent understanding of the ontology of injury and of related phenomena such as lesion, trauma, pain, and so forth. Further topics are listed [[Injury | here]].<br />
<br />
=='''Agenda: Friday Feb 24'''==<br />
<br />
09:00-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health.<br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
*'''Justin Reese:''' Unsupervised machine learning to define subtypes of long COVID using the Human Phenotype Ontology. [[Authors | Co-authors]] <br />
<br />
'''Abstract:''' Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterized by a wide range of manifestations that are difficult to analyze computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.<br />
<br />
We present a method for defining subtypes of long COVID by computationally modeling PASC phenotype data and applying unsupervised machine learning. We extracted Long COVID phenotype data expressed as HPO (Human Phenotype Ontology) terms from electronic healthcare records (EHRs), and then used semantic similarity of phenotype data to calculate a matrix of pairwise patient similarity. This matrix was then used to clustered patients using unsupervised machine learning (k means clustering).<br />
<br />
We identified six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centers to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems.<br />
<br />
Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.<br />
<br />
12:30-01:30p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations of industry partners<br />
<br />
== '''Participants''' ==<br />
<br />
* Barry Smith, co-organizer, University at Buffalo<br />
* William Hogan, co-organizer, University of Florida<br />
* Jihad Obeid, co-organizer, Medical University of South Carolina<br />
<br />
* Hamilton Baker, Medical University of South Carolina<br />
* Jobst Landgrebe, Cognotekt, Cologne<br />
* Tom Hutchinson, University of Pennsylvania, Institute for BioInformatics<br />
* Anna Maria Masci, National Institute of Environmental Health Sciences (NIEHS)<br />
* Richard Ohrbach, University at Buffalo<br />
* Justin Reese, Lawrence Berkeley National Laboratory<br />
* Richard Scheuermann, J. Craig Venter Institute<br />
* Matt Skowronek, TriNetX<br />
* Yonghui Wu, University of Florida</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73345CTS Ontology Workshop 20232023-01-30T17:47:47Z<p>Jihad: /* Sponsors */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
[[File:Chs23sm.jpg|link=|right|alt=Charleston]]<br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
<br />
<font size="+1">Special Focus Areas:</font><br />
<br />
* Social Determinants of Health<br />
* Mental health<br />
* EHR across the lifespan<br />
<br />
== '''Organizers''' ==<br />
<br />
Workshop Co-organizers:<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
<font size="+1">CTSOG Co-chairs:</font><br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* '''Medical University of South Carolina, the Biomedical Informatics Center''' and:<br />
<br />
[[File:Markiii-logo_160.png|link=https://www.markiiisys.com/|Mark III Systems]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:Nvidia-logo_180.png|link=https://www.nvidia.com/en-us/|NVIDIA]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:TriNetX-Logo-Horizontal-White_180.jpg|link=http://trinetx.com/|TriNetX]]<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Agenda: Thursday Feb 23rd'''==<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
*'''Introduction to the meeting, with brief presentations on:<br />
:The Clinical and Translational Science ontology landscope in 2023<br />
:Opportunities for AI in medicine<br />
:Will AI make ontologies redundant?<br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
*'''Justin Reese:''' Unsupervised machine learning to define subtypes of long COVID using the Human Phenotype Ontology. [[Authors]] <br />
<br />
'''Abstract:''' Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterized by a wide range of manifestations that are difficult to analyze computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.<br />
<br />
We present a method for defining subtypes of long COVID by computationally modeling PASC phenotype data and applying unsupervised machine learning. We extracted Long COVID phenotype data expressed as HPO (Human Phenotype Ontology) terms from electronic healthcare records (EHRs), and then used semantic similarity of phenotype data to calculate a matrix of pairwise patient similarity. This matrix was then used to clustered patients using unsupervised machine learning (k means clustering).<br />
<br />
We identified six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centers to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems.<br />
<br />
Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.<br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
*'''Barry Smith and Jobst Landgrebe:''' Discussion on AI and Medicine, with a special focus on ChatGPT<br />
<br />
'''Abstract: '''How are we to explain the peculiar tendency of ChatGPT to throw up what are called ‘hallucinations’? To answer this question we will draw on our book ''[https://buffalo.app.box.com/v/AI-Without-Fear Why Machines Will Never Rule the World]'', whose core thesis is that an artificial intelligence that could equal or exceed human intelligence—sometimes called "artificial general intelligence" (AGI)—is for mathematical reasons impossible. The argument for this thesis rests on the fact that (1.) human intelligence is a capability of a complex dynamic system—the human brain and central nervous system, and (2.) systems of this sort (like all organic systems) cannot be modelled mathematically in a way that would allow the models to operate inside a computer. We survey on this basis the potential of AI in the future of clinical and translational research.<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations<br />
<br />
*'''Yonghui Wu:''' A large language model for electronic health records.<br />
<br />
'''Abstract:''' Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
==Agenda: Friday Feb 24th==<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health.<br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
*'''Richard Ohrbach and Barry Smith:''' Defining 'Injury'<br />
<br />
'''Abstract:''' This session is intended as a contribution to the ontology component of a project for investigating injury and pain. It responds to the NIH [https://www.nidcr.nih.gov/grants-funding/funding-priorities/future-research-initiatives/tmd-collaborative-improving-patientcentered-translational-research-tmd-impact TMD IMPACT Collaborative for IMproving PAtient-Centered Translational Research]. SNOMED-CT defines an injury as a 'disorder resulting from physical damage to the body'. The WHO defines an injury as a 'bodily lesion at the organic level, resulting from acute exposure to energy (mechanical, thermal, electrical, chemical or radiant), in amounts that exceed the threshold of physiological tolerance.' We will explore these and other definitions with a view to establishing a more coherent understanding of the ontology of injury and of related phenomena such as lesion, trauma, pain, and so forth. Further topics are listed [[Injury | here]].<br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations<br />
<br />
== '''Participants''' ==<br />
<br />
* Barry Smith, co-organizer, University at Buffalo<br />
* William Hogan, co-organizer, University of Florida<br />
* Jobst Landgrebe, Cognotekt, Cologne<br />
* Jihad Obeid, co-organizer, Medical University of South Carolina<br />
* Hamilton Baker, Medical University of South Carolina<br />
* Jobst Landgrebe, University at Buffalo<br />
* Anna Maria Masci, National Institute of Environmental Health Sciences (NIEHS)<br />
* Richard Ohrbach, University at Buffalo<br />
* Justin Reese, Lawrence Berkeley National Laboratory<br />
* Yonghui Wu, University of Florida</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=File:TriNetX-Logo-Horizontal-White_180.jpg&diff=73344File:TriNetX-Logo-Horizontal-White 180.jpg2023-01-30T17:47:30Z<p>Jihad: </p>
<hr />
<div></div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=File:TriNetX-Logo-Horizontal-White.jpg&diff=73343File:TriNetX-Logo-Horizontal-White.jpg2023-01-30T17:44:41Z<p>Jihad: </p>
<hr />
<div></div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73326CTS Ontology Workshop 20232023-01-27T19:37:00Z<p>Jihad: </p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
[[File:Chs23sm.jpg|link=|right|alt=Charleston]]<br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
<br />
<font size="+1">Special Focus Areas:</font><br />
<br />
* Social Determinants of Health<br />
* Mental health<br />
* EHR across the lifespan<br />
<br />
== '''Organizers''' ==<br />
<br />
Workshop Co-organizers:<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
<font size="+1">CTSOG Co-chairs:</font><br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* '''Medical University of South Carolina, the Biomedical Informatics Center''' and:<br />
<br />
[[File:Markiii-logo_160.png|link=https://www.markiiisys.com/|Mark III Systems]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:Nvidia-logo_180.png|link=https://www.nvidia.com/en-us/|NVIDIA]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:TriNetX.png|link=http://trinetx.com/|TriNetX]]<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Speakers'''==<br />
<br />
*'''Barry Smith:''' The Current State of Ontology In Clinical and Translational Science.<br />
*'''Justin Reese:''' Unsupervised machine learning to define subtypes of long COVID using the Human Phenotype Ontology. '''Authors:''' Justin T. Reese, Hannah Blau, Elena Casiraghi, Timothy Bergquist, Johanna J. Loomba, Tiffany J. Callahan, Bryan Laraway, Corneliu Antonescu, Ben Coleman, Michael Gargano, Kenneth J. Wilkins, Luca Cappelletti, Tommaso Fontana, Nariman Ammar, Blessy Antony, T.M. Murali, J. Harry Caufield, Guy Karlebach, Julie A. McMurry, Andrew Williams, Richard Moffitt, Jineta Banerjee, Anthony E. Solomonides, Hannah Davis, Kristin Kostka, Giorgio Valentini, David Sahner, Christopher G. Chute, Charisse Madlock-Brown, Melissa A. Haendel, Peter N. Robinson.<br />
<br />
'''Abstract:''' Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterized by a wide range of manifestations that are difficult to analyze computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.<br />
<br />
We present a method for defining subtypes of long COVID by computationally modeling PASC phenotype data and applying unsupervised machine learning. We extracted Long COVID phenotype data expressed as HPO (Human Phenotype Ontology) terms from electronic healthcare records (EHRs), and then used semantic similarity of phenotype data to calculate a matrix of pairwise patient similarity. This matrix was then used to clustered patients using unsupervised machine learning (k means clustering).<br />
<br />
We identified six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centers to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems.<br />
<br />
Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.<br />
<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health.<br />
*'''Yonghui Wu:''' A large language model for electronic health records.<br />
<br />
'''Abstract:''' Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
== '''Agenda'''==<br />
<br />
'''Thursday Feb 23rd'''<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations<br />
<br />
'''Friday Feb 24th'''<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations<br />
<br />
== '''Participants''' ==<br />
<br />
* Barry Smith, co-organizer, University at Buffalo<br />
* William Hogan, co-organizer, University of Florida<br />
* Jihad Obeid, co-organizer, Medical University of South Carolina<br />
<br />
* Hamilton Baker, Medical University of South Carolina<br />
* Jobst Landgrebe, University at Buffalo<br />
* Anna Maria Masci, National Institute of Environmental Health Sciences (NIEHS)<br />
* Justin Reese, Lawrence Berkeley National Laboratory<br />
* Yonghui Wu, University of Florida</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=File:Chs23sm.jpg&diff=73325File:Chs23sm.jpg2023-01-27T19:26:30Z<p>Jihad: </p>
<hr />
<div></div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73324CTS Ontology Workshop 20232023-01-26T22:22:54Z<p>Jihad: /* Sponsors */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
<br />
<font size="+1">Special Focus Areas:</font><br />
<br />
* Social Determinants of Health<br />
* Mental health<br />
* EHR across the lifespan<br />
<br />
== '''Organizers''' ==<br />
<br />
Workshop Co-organizers:<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
<font size="+1">CTSOG Co-chairs:</font><br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* '''Medical University of South Carolina, the Biomedical Informatics Center''' and:<br />
<br />
[[File:Markiii-logo_160.png|link=https://www.markiiisys.com/|Mark III Systems]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:Nvidia-logo_180.png|link=https://www.nvidia.com/en-us/|NVIDIA]] &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp;<br />
[[File:TriNetX.png|link=http://trinetx.com/|TriNetX]]<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Speakers'''==<br />
<br />
*'''Barry Smith:''' The Current State of Ontology In Clinical and Translational Science.<br />
*'''Justin Reese:''' Unsupervised machine learning to define subtypes of long COVID using the Human Phenotype Ontology. '''Authors:''' Justin T. Reese, Hannah Blau, Elena Casiraghi, Timothy Bergquist, Johanna J. Loomba, Tiffany J. Callahan, Bryan Laraway, Corneliu Antonescu, Ben Coleman, Michael Gargano, Kenneth J. Wilkins, Luca Cappelletti, Tommaso Fontana, Nariman Ammar, Blessy Antony, T.M. Murali, J. Harry Caufield, Guy Karlebach, Julie A. McMurry, Andrew Williams, Richard Moffitt, Jineta Banerjee, Anthony E. Solomonides, Hannah Davis, Kristin Kostka, Giorgio Valentini, David Sahner, Christopher G. Chute, Charisse Madlock-Brown, Melissa A. Haendel, Peter N. Robinson.<br />
<br />
'''Abstract:''' Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterized by a wide range of manifestations that are difficult to analyze computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.<br />
<br />
We present a method for defining subtypes of long COVID by computationally modeling PASC phenotype data and applying unsupervised machine learning. We extracted Long COVID phenotype data expressed as HPO (Human Phenotype Ontology) terms from electronic healthcare records (EHRs), and then used semantic similarity of phenotype data to calculate a matrix of pairwise patient similarity. This matrix was then used to clustered patients using unsupervised machine learning (k means clustering).<br />
<br />
We identified six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centers to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems.<br />
<br />
Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.<br />
<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health.<br />
*'''Yonghui Wu:''' A large language model for electronic health records.<br />
<br />
'''Abstract:''' Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
== '''Agenda'''==<br />
<br />
'''Thursday Feb 23rd'''<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations<br />
<br />
'''Friday Feb 24th'''<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations<br />
<br />
== '''Participants''' ==<br />
<br />
* Barry Smith, co-organizer, University at Buffalo<br />
* William Hogan, co-organizer, University of Florida<br />
* Jihad Obeid, co-organizer, Medical University of South Carolina<br />
<br />
* Hamilton Baker, Medical University of South Carolina<br />
* Jobst Landgrebe, University at Buffalo<br />
* Anna Maria Masci, National Institute of Environmental Health Sciences (NIEHS)<br />
* Justin Reese, Lawrence Berkeley National Laboratory<br />
* Yonghui Wu, University of Florida</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=File:Nvidia-logo_180.png&diff=73323File:Nvidia-logo 180.png2023-01-26T22:20:29Z<p>Jihad: </p>
<hr />
<div></div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=File:Markiii-logo_160.png&diff=73322File:Markiii-logo 160.png2023-01-26T22:16:39Z<p>Jihad: Markiii-logo</p>
<hr />
<div>== Summary ==<br />
Markiii-logo</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73315CTS Ontology Workshop 20232023-01-23T20:14:35Z<p>Jihad: </p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
<br />
<font size="+1">Special Focus Areas:</font><br />
<br />
* Social Determinants of Health<br />
* Mental health<br />
* EHR across the lifespan<br />
<br />
== '''Organizers''' ==<br />
<br />
Workshop Co-organizers:<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
<font size="+1">CTSOG Co-chairs:</font><br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* Medical University of South Carolina and the Biomedical Informatics Center<br />
* Others (coming soon)<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Speakers'''==<br />
<br />
*'''Barry Smith:''' The Current State of Ontology In Clinical and Translational Science.<br />
*'''Justin Reese:''' Unsupervised machine learning to define subtypes of long COVID using the Human Phenotype Ontology. '''Authors:''' Justin T. Reese, Hannah Blau, Elena Casiraghi, Timothy Bergquist, Johanna J. Loomba, Tiffany J. Callahan, Bryan Laraway, Corneliu Antonescu, Ben Coleman, Michael Gargano, Kenneth J. Wilkins, Luca Cappelletti, Tommaso Fontana, Nariman Ammar, Blessy Antony, T.M. Murali, J. Harry Caufield, Guy Karlebach, Julie A. McMurry, Andrew Williams, Richard Moffitt, Jineta Banerjee, Anthony E. Solomonides, Hannah Davis, Kristin Kostka, Giorgio Valentini, David Sahner, Christopher G. Chute, Charisse Madlock-Brown, Melissa A. Haendel, Peter N. Robinson.<br />
<br />
'''Abstract:''' Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterized by a wide range of manifestations that are difficult to analyze computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.<br />
<br />
We present a method for defining subtypes of long COVID by computationally modeling PASC phenotype data and applying unsupervised machine learning. We extracted Long COVID phenotype data expressed as HPO (Human Phenotype Ontology) terms from electronic healthcare records (EHRs), and then used semantic similarity of phenotype data to calculate a matrix of pairwise patient similarity. This matrix was then used to clustered patients using unsupervised machine learning (k means clustering).<br />
<br />
We identified six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centers to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems.<br />
<br />
Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.<br />
<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health.<br />
*'''Yonghui Wu:''' A large language model for electronic health records.<br />
<br />
'''Abstract:''' Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
== '''Agenda'''==<br />
<br />
'''Thursday Feb 23rd'''<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations<br />
<br />
'''Friday Feb 24th'''<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations<br />
<br />
== '''Participants''' ==<br />
<br />
* Barry Smith, co-organizer, University at Buffalo<br />
* William Hogan, co-organizer, University of Florida<br />
* Jihad Obeid, co-organizer, Medical University of South Carolina<br />
<br />
* Hamilton Baker, Medical University of South Carolina<br />
* Jobst Landgrebe, University at Buffalo<br />
* Anna Maria Masci, National Institute of Environmental Health Sciences (NIEHS)<br />
* Justin Reese, Lawrence Berkeley National Laboratory<br />
* Yonghui Wu, University of Florida</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73304CTS Ontology Workshop 20232023-01-20T20:32:10Z<p>Jihad: /* Participants */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
<br />
<font size="+1">Special Focus Areas:</font><br />
<br />
* Social Determinants of Health<br />
* Mental health<br />
* EHR across the lifespan<br />
<br />
== '''Organizers''' ==<br />
<br />
Workshop Co-organizers:<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
<font size="+1">CTSOG Co-chairs:</font><br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* Medical University of South Carolina and the Biomedical Informatics Center<br />
* Others (coming soon)<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
*'''Barry Smith:''' The Current State of Ontology In Clinical and Translational Science.<br />
*'''Justin Reese:''' Unsupervised machine learning to define subtypes of long COVID using the Human Phenotype Ontology. '''Authors:''' Justin T. Reese, Hannah Blau, Elena Casiraghi, Timothy Bergquist, Johanna J. Loomba, Tiffany J. Callahan, Bryan Laraway, Corneliu Antonescu, Ben Coleman, Michael Gargano, Kenneth J. Wilkins, Luca Cappelletti, Tommaso Fontana, Nariman Ammar, Blessy Antony, T.M. Murali, J. Harry Caufield, Guy Karlebach, Julie A. McMurry, Andrew Williams, Richard Moffitt, Jineta Banerjee, Anthony E. Solomonides, Hannah Davis, Kristin Kostka, Giorgio Valentini, David Sahner, Christopher G. Chute, Charisse Madlock-Brown, Melissa A. Haendel, Peter N. Robinson.<br />
<br />
'''Abstract:''' Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterized by a wide range of manifestations that are difficult to analyze computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.<br />
<br />
We present a method for defining subtypes of long COVID by computationally modeling PASC phenotype data and applying unsupervised machine learning. We extracted Long COVID phenotype data expressed as HPO (Human Phenotype Ontology) terms from electronic healthcare records (EHRs), and then used semantic similarity of phenotype data to calculate a matrix of pairwise patient similarity. This matrix was then used to clustered patients using unsupervised machine learning (k means clustering).<br />
<br />
We identified six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centers to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems.<br />
<br />
Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.<br />
<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health.<br />
*'''Yonghui Wu:''' A large language model for electronic health records.<br />
<br />
'''Abstract:''' Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
== '''Agenda'''==<br />
<br />
'''Thursday Feb 23rd'''<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations<br />
<br />
'''Friday Feb 24th'''<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations<br />
<br />
== '''Participants''' ==<br />
<br />
* Barry Smith, co-organizer, University at Buffalo<br />
* William Hogan, co-organizer, University of Florida<br />
* Jihad Obeid, co-organizer, Medical University of South Carolina<br />
<br />
* Hamilton Baker, Medical University of South Carolina<br />
* Jobst Landgrebe, University at Buffalo<br />
* Anna Maria Masci, National Institute of Environmental Health Sciences (NIEHS)<br />
* Justin Reese, Lawrence Berkeley National Laboratory<br />
* Yonghui Wu, University of Florida</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73303CTS Ontology Workshop 20232023-01-20T20:27:54Z<p>Jihad: /* Speakers */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
<br />
<font size="+1">Special Focus Areas:</font><br />
<br />
* Social Determinants of Health<br />
* Mental health<br />
* EHR across the lifespan<br />
<br />
== '''Organizers''' ==<br />
<br />
Workshop Co-organizers:<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
<font size="+1">CTSOG Co-chairs:</font><br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* Medical University of South Carolina and the Biomedical Informatics Center<br />
* Others (coming soon)<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
*'''Barry Smith:''' The Current State of Ontology In Clinical and Translational Science.<br />
*'''Justin Reese:''' Unsupervised machine learning to define subtypes of long COVID using the Human Phenotype Ontology. '''Authors:''' Justin T. Reese, Hannah Blau, Elena Casiraghi, Timothy Bergquist, Johanna J. Loomba, Tiffany J. Callahan, Bryan Laraway, Corneliu Antonescu, Ben Coleman, Michael Gargano, Kenneth J. Wilkins, Luca Cappelletti, Tommaso Fontana, Nariman Ammar, Blessy Antony, T.M. Murali, J. Harry Caufield, Guy Karlebach, Julie A. McMurry, Andrew Williams, Richard Moffitt, Jineta Banerjee, Anthony E. Solomonides, Hannah Davis, Kristin Kostka, Giorgio Valentini, David Sahner, Christopher G. Chute, Charisse Madlock-Brown, Melissa A. Haendel, Peter N. Robinson.<br />
<br />
'''Abstract:''' Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterized by a wide range of manifestations that are difficult to analyze computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.<br />
<br />
We present a method for defining subtypes of long COVID by computationally modeling PASC phenotype data and applying unsupervised machine learning. We extracted Long COVID phenotype data expressed as HPO (Human Phenotype Ontology) terms from electronic healthcare records (EHRs), and then used semantic similarity of phenotype data to calculate a matrix of pairwise patient similarity. This matrix was then used to clustered patients using unsupervised machine learning (k means clustering).<br />
<br />
We identified six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centers to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems.<br />
<br />
Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.<br />
<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health.<br />
*'''Yonghui Wu:''' A large language model for electronic health records.<br />
<br />
'''Abstract:''' Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
== '''Agenda'''==<br />
<br />
'''Thursday Feb 23rd'''<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations<br />
<br />
'''Friday Feb 24th'''<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations<br />
<br />
== '''Participants''' ==<br />
<br />
*Barry Smith, co-organizer, University at Buffalo<br />
*William Hogan, co-organizer, University of Florida<br />
*Jihad Obeid, co-organizer, Medical University of South Carolina<br />
<br />
*Jobst Landgrebe, University at Buffalo</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73277CTS Ontology Workshop 20232023-01-18T19:29:22Z<p>Jihad: /* Speakers */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
<br />
<font size="+1">Special Focus Areas:</font><br />
<br />
* Social Determinants of Health<br />
* Mental health<br />
* EHR across the lifespan<br />
<br />
== '''Organizers''' ==<br />
<br />
Workshop Co-organizers:<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
<font size="+1">CTSOG Co-chairs:</font><br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* Medical University of South Carolina and the Biomedical Informatics Center<br />
* Others (coming soon)<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Speakers'''==<br />
<br />
*'''Barry Smith:''' The Current State of Ontology In Clinical and Translational Science.<br />
*'''Justin Reese:''' Leveraging the Human Phenotype Ontology for an ML Approach to Identify Long COVID.<br />
*'''William Hogan:''' Semantic Representation of Occupations as Social Determinants of Health.<br />
*'''Yonghui Wu:''' A large language model for electronic health records. '''Abstract:''' Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. We develop from scratch a large clinical language model—GatorTron—using >90 billion words of text (including >82 billion words of de-identified clinical text). GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve five clinical NLP tasks, which can be applied to medical AI systems to improve healthcare delivery.<br />
<br />
== '''Agenda'''==<br />
<br />
'''Thursday Feb 23rd'''<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations<br />
<br />
'''Friday Feb 24th'''<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations<br />
<br />
== '''Participants''' ==<br />
<br />
*Barry Smith, co-organizer, University at Buffalo<br />
*William Hogan, co-organizer, University of Florida<br />
*Jihad Obeid, co-organizer, Medical University of South Carolina<br />
<br />
*Jobst Landgrebe, University at Buffalo</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73270CTS Ontology Workshop 20232023-01-11T21:49:26Z<p>Jihad: </p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
<br />
<font size="+1">Special Focus Areas:</font><br />
<br />
* Social Determinants of Health<br />
* Mental health<br />
* EHR across the lifespan<br />
<br />
== '''Organizers''' ==<br />
<br />
Workshop Co-organizers:<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
<font size="+1">CTSOG Co-chairs:</font><br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* Medical University of South Carolina and the Biomedical Informatics Center<br />
* Others (coming soon)<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Speakers'''==<br />
<br />
*Barry Smith: The Current State of Ontology In Clinical and Translational Science.<br />
*Justin Reese: Leveraging the Human Phenotype Ontology for an ML Approach to Identify Long COVID.<br />
*William Hogan: Semantic Representation of Occupations as Social Determinants of Health.<br />
*TBD: Large Language Models and Their Impact on Medicine and Ontology Development.<br />
<br />
== '''Agenda'''==<br />
<br />
'''Thursday Feb 23rd'''<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations<br />
<br />
'''Friday Feb 24th'''<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations<br />
<br />
== '''Participants''' ==<br />
<br />
*Barry Smith, co-organizer, University at Buffalo<br />
*William Hogan, co-organizer, University of Florida<br />
*Jihad Obeid, co-organizer, Medical University of South Carolina<br />
<br />
*Jobst Landgrebe, University at Buffalo</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73269CTS Ontology Workshop 20232023-01-09T21:35:04Z<p>Jihad: /* Venue */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
<br />
<font size="+1">Special Focus Areas:</font><br />
<br />
* Social Determinants of Health<br />
* Mental health<br />
* EHR across the lifespan<br />
<br />
== '''Organizers''' ==<br />
<br />
Workshop Co-organizers:<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
<font size="+1">CTSOG Co-chairs:</font><br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* Medical University of South Carolina and the Biomedical Informatics Center<br />
* Others (coming soon)<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 10 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Agenda'''==<br />
<br />
'''Thursday Feb 23rd'''<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations<br />
<br />
'''Friday Feb 24th'''<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations<br />
<br />
== '''Participants''' ==<br />
<br />
*Barry Smith, co-organizer, University at Buffalo<br />
*William Hogan, co-organizer, University of Florida<br />
*Jihad Obeid, co-organizer, Medical University of South Carolina<br />
<br />
*Jobst Landgrebe, University at Buffalo</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=File:Ctso_2023_map_sm.png&diff=73268File:Ctso 2023 map sm.png2023-01-09T20:36:41Z<p>Jihad: </p>
<hr />
<div></div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73267CTS Ontology Workshop 20232023-01-09T20:30:00Z<p>Jihad: /* Venue */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
<br />
<font size="+1">Special Focus Areas:</font><br />
<br />
* Social Determinants of Health<br />
* Mental health<br />
* EHR across the lifespan<br />
<br />
== '''Organizers''' ==<br />
<br />
Workshop Co-organizers:<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
<font size="+1">CTSOG Co-chairs:</font><br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* Medical University of South Carolina and the Biomedical Informatics Center<br />
* Others (coming soon)<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
'''Directions:''' From Courtyard Marriott on Lockwood<br />
<br />
* 10 minute walk, 0.4 miles, safe crosswalks along the route<br />
* Hotel shuttle – runs periodically, need to ask hotel desk for timing details.<br />
<br />
'''Paid parking:''' <br />
<br />
* Parking deck at 21 WestEdge (building before 22 WestEdge, with Publix on bottom floor)<br />
* App-based parking next to 22 WestEdge building (signage with instructions posted onsite)<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Agenda'''==<br />
<br />
'''Thursday Feb 23rd'''<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations<br />
<br />
'''Friday Feb 24th'''<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations<br />
<br />
== '''Participants''' ==<br />
<br />
*Barry Smith, co-organizer, University at Buffalo<br />
*William Hogan, co-organizer, University of Florida<br />
*Jihad Obeid, co-organizer, Medical University of South Carolina<br />
<br />
*Jobst Landgrebe, University at Buffalo</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73266CTS Ontology Workshop 20232023-01-09T20:20:40Z<p>Jihad: /* Hotel */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
<br />
<font size="+1">Special Focus Areas:</font><br />
<br />
* Social Determinants of Health<br />
* Mental health<br />
* EHR across the lifespan<br />
<br />
== '''Organizers''' ==<br />
<br />
Workshop Co-organizers:<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
<font size="+1">CTSOG Co-chairs:</font><br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* Medical University of South Carolina and the Biomedical Informatics Center<br />
* Others (coming soon)<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670450257241&key=GRP&app=resvlink Feb. 22-23 reservation link]<br />
<br />
[https://www.marriott.com/event-reservations/reservation-link.mi?id=1670451475075&key=GRP&app=resvlink Feb. 24-25 reservation link] (use this link if you are extending your stay in Charleston through the weekend)<br />
<br />
If you have any issues registering, please call the hotel directly at 843-722-7229 and mention you are visiting for the CTSA Ontology meeting.<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Agenda'''==<br />
<br />
'''Thursday Feb 23rd'''<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations<br />
<br />
'''Friday Feb 24th'''<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations<br />
<br />
== '''Participants''' ==<br />
<br />
*Barry Smith, co-organizer, University at Buffalo<br />
*William Hogan, co-organizer, University of Florida<br />
*Jihad Obeid, co-organizer, Medical University of South Carolina<br />
<br />
*Jobst Landgrebe, University at Buffalo</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73263CTS Ontology Workshop 20232023-01-06T15:04:07Z<p>Jihad: /* Sponsors */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
<br />
<font size="+1">Special Focus Areas:</font><br />
<br />
* Social Determinants of Health<br />
* Mental health<br />
* EHR across the lifespan<br />
<br />
== '''Organizers''' ==<br />
<br />
Workshop Co-organizers:<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
<font size="+1">CTSOG Co-chairs:</font><br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* Medical University of South Carolina and the Biomedical Informatics Center<br />
* Others (coming soon)<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Agenda'''==<br />
<br />
'''Thursday Feb 23rd'''<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations<br />
<br />
'''Friday Feb 24th'''<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations<br />
<br />
== '''Participants''' ==<br />
<br />
*Barry Smith, co-organizer, University at Buffalo<br />
*William Hogan, co-organizer, University of Florida<br />
*Jihad Obeid, co-organizer, Medical University of South Carolina<br />
<br />
*Jobst Landgrebe, University at Buffalo</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73262CTS Ontology Workshop 20232023-01-06T15:03:29Z<p>Jihad: /* Sponsors */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
<br />
<font size="+1">Special Focus Areas:</font><br />
<br />
* Social Determinants of Health<br />
* Mental health<br />
* EHR across the lifespan<br />
<br />
== '''Organizers''' ==<br />
<br />
Workshop Co-organizers:<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
<font size="+1">CTSOG Co-chairs:</font><br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* Medical University of South Carolina and the Biomedical Informatics Center<br />
* Others (Coming soon)<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Agenda'''==<br />
<br />
'''Thursday Feb 23rd'''<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations<br />
<br />
'''Friday Feb 24th'''<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations<br />
<br />
== '''Participants''' ==<br />
<br />
*Barry Smith, co-organizer, University at Buffalo<br />
*William Hogan, co-organizer, University of Florida<br />
*Jihad Obeid, co-organizer, Medical University of South Carolina<br />
<br />
*Jobst Landgrebe, University at Buffalo</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73261CTS Ontology Workshop 20232023-01-05T22:07:07Z<p>Jihad: </p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[http://ncorwiki.buffalo.edu/index.php/Clinical_and_Translational_Science_Ontology_Group Clinical and Translational Science Ontology Group (CTSOG)]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
<br />
<font size="+1">Special Focus Areas:</font><br />
<br />
* Social Determinants of Health<br />
* Mental health<br />
* EHR across the lifespan<br />
<br />
== '''Organizers''' ==<br />
<br />
Workshop Co-organizers:<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
<font size="+1">CTSOG Co-chairs:</font><br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* Medical University of South Carolina and the Biomedical Informatics Center<br />
* Others<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Agenda'''==<br />
<br />
'''Thursday Feb 23rd'''<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations<br />
<br />
'''Friday Feb 24th'''<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations<br />
<br />
== '''Participants''' ==<br />
<br />
*Barry Smith, co-organizer, University at Buffalo<br />
*William Hogan, co-organizer, University of Florida<br />
*Jihad Obeid, co-organizer, Medical University of South Carolina<br />
<br />
*Jobst Landgrebe, University at Buffalo</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73260CTS Ontology Workshop 20232023-01-05T21:59:02Z<p>Jihad: </p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[[Clinical and Translational Science Ontology Group (CTSOG)]]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
<br />
<font size="+1">Special Focus Areas:</font><br />
<br />
* Social Determinants of Health<br />
* Mental health<br />
* EHR across the lifespan<br />
<br />
== '''Organizers''' ==<br />
<br />
Workshop Co-organizers:<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
<font size="+1">CTSOG Co-chairs:</font><br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* Medical University of South Carolina and the Biomedical Informatics Center<br />
* Others<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Agenda'''==<br />
<br />
'''Thursday Feb 23rd'''<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations<br />
<br />
'''Friday Feb 24th'''<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations<br />
<br />
== '''Participants''' ==<br />
<br />
*Barry Smith, co-organizer, University at Buffalo<br />
*William Hogan, co-organizer, University of Florida<br />
*Jihad Obeid, co-organizer, Medical University of South Carolina<br />
<br />
*Jobst Landgrebe, University at Buffalo</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73259CTS Ontology Workshop 20232023-01-05T21:56:32Z<p>Jihad: /* Schedule */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[[Clinical and Translational Science Ontology Group (CTSOG)]]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
<br />
<font size="+1">Special Focus Areas:</font><br />
<br />
* Social Determinants of Health<br />
* Mental health<br />
* EHR across the lifespan<br />
<br />
== '''Organizers''' ==<br />
<br />
Workshop Co-organizers:<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
<font size="+1">CTSOG Co-chairs:</font><br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* Medical University of South Carolina and the Biomedical Informatics Center<br />
* Others<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Agenda'''==<br />
<br />
'''Thursday Feb 23rd'''<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations<br />
<br />
'''Friday Feb 24th'''<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations</div>Jihadhttps://ncorwiki.buffalo.edu/index.php?title=CTS_Ontology_Workshop_2023&diff=73258CTS Ontology Workshop 20232023-01-05T21:55:42Z<p>Jihad: /* Registration */</p>
<hr />
<div><font size="+3">Ontologies, AI and Electronic Health Records</font> <br />
<br />
More about the '''[[Clinical and Translational Science Ontology Group (CTSOG)]]''' and previous meetings.<br />
<br />
<font size="+2">Feb 23 - 24, 2023 - Charleston, SC</font><br />
<br />
__TOC__<br />
<br />
== '''Background''' ==<br />
<br />
'''The Clinical and Translational Science Ontology Group (CTSOG)''' invites you to join us February 23-24, 2023, in Charleston, SC to discuss the role of ontologies in improving electronic health records (EHR) systems by advancing semantic interoperability, translational research, and artificial intelligence (AI). As health data increases in volume and complexity, and artificial intelligence applications gain momentum, the need for careful planning and interoperability becomes more critical. The purpose of this workshop is to explore new paradigms in EHRs, by examining successes and failures of ontologies and AI in different areas of biomedicine and their role in equitable healthcare. <br />
<br />
== '''Themes''' ==<br />
<br />
* Improving the EHR with ontologies and with AI<br />
* The functions of the EHR and other healthcare documents<br />
<br />
<font size="+1">Special Focus Areas:</font><br />
<br />
* Social Determinants of Health<br />
* Mental health<br />
* EHR across the lifespan<br />
<br />
== '''Organizers''' ==<br />
<br />
Workshop Co-organizers:<br />
<br />
Bill Hogan, Jihad Obeid, Barry Smith<br />
<br />
<font size="+1">CTSOG Co-chairs:</font><br />
<br />
Bill Hogan (University of Florida College of Medicine, Gainesville, FL), hoganwr@ufl.edu<br />
<br />
Barry Smith (University at Buffalo, Buffalo, NY), phismith@buffalo.edu<br />
<br />
== '''Sponsors''' ==<br />
<br />
* Medical University of South Carolina and the Biomedical Informatics Center<br />
* Others<br />
<br />
== '''Venue'''==<br />
<br />
Biomedical Informatics Center<br />
<br />
Medical University of South Carolina<br />
<br />
'''Address:''' 22 Westedge St, Charleston, SC 29403<br />
<br />
== '''Hotel'''==<br />
<br />
'''Courtyard by Marriott Charleston Waterfront'''<br />
<br />
'''Address:''' 35 Lockwood Dr, Charleston, SC 29401<br />
<br />
'''Phone:''' (843) 722-7229<br />
<br />
=='''Registration'''== <br />
<br />
[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y Registration is free].<br />
<br />
However, '''[https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y registration is required]''' for planning purposes. Please register for the workshop using [https://redcap.musc.edu/surveys/?s=W7FNH78FFHCRNN9Y this form].<br />
<br />
== '''Schedule'''==<br />
<br />
'''Thursday Feb 23rd'''<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations<br />
<br />
'''Friday Feb 24th'''<br />
<br />
'''09:00'''-09:30a Breakfast and Networking<br />
<br />
09:30-10:45a Working sessions/Presentations <br />
<br />
--break--<br />
<br />
11:00-12:30p Working sessions/Presentations <br />
<br />
12:30-01:00p Networking and Working Lunch <br />
<br />
1:00-03:15p Working sessions/Presentations<br />
<br />
--break--<br />
<br />
03:30-'''5:00p''' Working sessions/Presentations</div>Jihad