Abstract: Elkin

Let’s say we are interested in the following research question:

Do patients who have Rosacia have a higher risk of Obstructive Sleep Apnea than the general population?

To answer this question we need to know:

1. Number of Patients in total who do not have Rosacia

2. Number of Patients who do not have Rosacia and have OSA

A. Number of Patients with Rosacia

B. Number of Patients with Rosacia and OSA

We then compare 2/1 with B/A using for example a Pearson chi square test.

The names would be converted to in this case SNOMED CT codes and matches with prior stored codes from the Health Record. This is a real example where people with Rosacia had a 2x rate of OSA.

Ontology-tagged clinical data can use the power of the ontological relations to a) use natural language queries into big data bases like EHRs and b) can use the hierarchies to find the transitive reflexive closure on subsumption so that the person asking clinical research questions does not need to know all the types of Cardiovascular disorder or anti-arrhythmic drugs as the ontology can assist. The normalization of the many ways to say the same thing to a single ontology term facilitates reliable information retrieval. Coupling these characteristics with an interface that allows subject-matter experts with no computer science background to ask complex questions of clinical data and get answers within seconds.