Computer-Based Medical Decision Making: From MYCIN to VM

We mentioned in the introduction to Chapter 5 that MYCIN provided a starting point for several additional research projects. The Ventilator Manager (VM) project of Larry Fagan had its beginnings in the MYCIN project but quickly diverged because of the dynamic nature of the intensive care unit (ICU) setting for which it was designed. MYCIN required a "snapshot" approach to patient assessment-temporal trends were poorly handled and advice was generally provided on the basis of a patient's situation at a single point in time. In dynamic settings like an ICU, however, decisions may be dependent on frequent sequential assessments of the patient's status. Fagan's work was accordingly also influenced by another earlier Stanford project known as HASP/SlAP (Nii et al., 1982). That system was not concerned with medical issues, but did develop iterative techniques for the ongoing analysis of signals. Although Fagan was a graduate student at Stanford at the time, much of his work was based at Pacific Medical Center in San Francisco. The director of the postsurgical ICU there, Dr. John Osborn, had developed an elaborate monitoring system that was in routine use. However, the amount of data generated was sometimes overwhelming, particularly for physicians in training. It was clear that there was expertise involved in learning how to interpret the data, and the idea developed to build an expert system that could monitor" the various physiological parameters and give advice accordingly. The following chapter discusses the resulting evolution from