Qualitative models in medical diagnosis

by Liliana Ironi, M. Stefanelli and G. Lanzola



in Artificial Intelligence in Medicine, 2, (1990), 85-101 (also in Deep Models for Medical Knowledge Engineering, E. Keravnou ed., (1992), Elsevier, 51-70).

ABSTRACT

Diagnostic systems based solely on associative knowledge are able of drawing accurate conclusions in acceptable times but they do not capture all the available medical knowledge. Some of this knowledge, even if incomplete, is sufficiently precise that qualitative models can be formulated. Aim of this paper is to discuss how qualitative models can be exploited in a medical diagnostic system. We present a system, called NEOANEMIA, integrating first-generation knowledge representation formalisms (frames and production rules) with qualitative pathophysiological models to diagnose haematologic disorders causing anaemia. To this purpose, qualitative models of iron metabolism, erythropoietin metabolism, red cell production and destruction have been formulated. Describing these models we will point out problems related to such a knowledge representation formalism. The key ideas of our work are: abducing diagnostic hypotheses from observed problem features, modeling pathophysiological systems with dynamic qualitative models, predicting pathophysiological behaviours by qualitative model simulation, comparing clinical observations against simulation results, and, when necessary, incrementally creating and testing multiple-diagnostic hypotheses. In this way the performance of a diagnostic expert system can be highly enhanced.



  • Back to publication list

    Liliana Ironi June 25th, 1996