Reference: Shwe, M.; Middleton, B.; Heckerman, D.; Henrion, M.; Horvitz, E.; Lehmann, H.; & Cooper, G. Probabilistic Diagnosis Using a Reformulation of the INTERNIST-1/QMR Knowledge Base I. The Probabilistic Model and Inference Algorithms. 1991.
Abstract: In this paper we report on the design and implementation of a two-level multiply-connected belief-network representation of the INTERNIST-1 knowledge base, a large heuristic knowledge base of internal medicine.In the belief- network representation, we use probabilities derived from the INTERNIST-1 disease profiles, INTERNIST-1 imports of findings, and from National Center for Health Statistics hospital discharge statistics. Using a stochastic simulation algorithm for inference on the belief network, we compared the performance of QMR to that of the probabilistic reformulation on cases abstracted from continuing medical education materials from Scientific American Medicine. In addition, we performed a sensitivity analysis on the probabilistic model and on the simulation algorithm.