A Model-Based Method for Computer-Aided Medical Decision-Making

While MYCIN and PIP were under development at Stanford and Tufts! M.I. T., a group of computer scientists at Rutgers University was developing a system to aid in the evaluation and treatment of patients with glaucoma. The group was led by Professor Casimir K ulikou1ski, a researcher with extensive background in mathematical and pattern-recognition approaches to computer-based medical decision making (Nordyke et al., 1971), working within the Rutgers Research Resource on Computers in Biomedicine headed by Professor Saul Amarel. Working collaboratively with Dr. Arin Safir, Professor of Ophthalmology, "who was then based at the Mt. Sinai School of Medicine in New York City, Kulikowski and Sholom Weiss (a graduate student at Rutgers who "went on to become a research scientist there) developed a method of computer-assisted medical decision making that was based on causal-associational network (CASNET) models of disease. Although the work was inspired by the glaucoma domain, the approach had general features that were later refined in the development of the EXPERT system-building tool (see Chapters 18 and 20). A CASNET model consists of three rnain components: observations of a patient, pathophysiological states, and disease classifications. As observations are recorded, they are associated with the appropriate intennediate states. These states, in turn, are typically causally related, thereby forming a network that summarizes the mechanisms of disease. It is these patterns of states in the network that are linked to individual disease classes. Strat-