ILIAD as an Expert Consultant to Teach Differential Diagnosis

Abstract

A program (ILIAD) is described which uses knowledge frames representing diseases encountered in internal medicine to teach medical students about differential diagnosis. ILIAD consultant utilizes a number of sophisticated inferencing mechanisms to mimic the strategy of a medical expert in working up a patient. Its knowledge is represented in both Bayesean and Boolean frames which permit use of sensitivities and specificities to describe the relationship of a disease to its manifestations and provide a basis for explaining its conclusions. In addition to differential diagnosis, ILIAD provides advice regarding the most appropriate information to seek at each stage of the workup. ILIAD's knowledge base is also used to simulate patient cases and evaluate the problem solving performance of medical apprentices. It is presently being used by third year medical clerks on the wards of three hospitals and beta tested at eight additional sites.