OntoQuest: a physician decision support system based on ontological queries of the hospital database.

OntoQuest is a physician decision support system that mines the hospital data base for previous decisions made in cases similar to the current one. For example, OntoQuest displays a list of the medications prescribed to similar historical patients from which the physician may compare his choice of medication for the current patient. This information retrieval is accomplished using ontological queries. Unlike a regular database query, an ontological query is able to account for semantic similarity between patients. To implement the ontological query, we propose a method for computing the ontological similarity between patients represented by sets of ICD-9 diagnoses. We have tested the OntoQuest prototype on a pilot data set of 2077 patients. Finally, we compare the OntoQuest performance to conventional database queries. We believe that OntoQuest can be extended to compare services, quality and outcomes among patient and provider groups.