Outline of an expert system for ECG diagnosis using fuzzy sets

摘要

Medical experts, in particular cardiologists, often use imprecise linguistic terms to describe the physical state under consideration. So they are talking about a low blood pressure, severe headache, a high amplitude. Moreover, the inference rules they use are also characterized by a shade of uncertainty which is reflected by rules of thumb of the form: if QRS duration is high then left ventricular hypertrophy is possible. The formalism of fuzzy set theory is extremely useful to handle these kinds of uncertainty and hence to offer a tool for automatic diagnostic analysis of ECGs and EEGs. In this paper we describe these techniques, creating in this way a starting point for an expert system for interpreting ECGs.