APPROXIMATE REASONING AND INTERPRETATION OF LABORATORY TESTS IN MEDICAL DIAGNOSTICS

Publisher Website
Google Scholar
Abstract
In many real-world domains, such as medicine, human knowledge is by nature imprecise. As a consequence, the expert systems oriented to these domains must have specific tools lo deal with the uncertainty. A commonly used approach is to equip the expert system with a computational capability to analyze the transmission of uncertainty from the premises lo the conclusion. The theory of fuzzy sets provides a systematic framework for dealing with fuzzy quantifiers, such as many, few, and most, and linguistic variables like normal, increased, tall, small, and old. In this framework the results of laboratory tests (we refer to the area of medicai diagnostics) that are precise may be interpreted in the terms of fuzzy propositions by means of membership functions. This approach permits the expert system to employ the same inference engine used to deal with the imprecise information provided by physicians. In this paper we describe a method for the construction of the membership functions using the statistical data and the range of normal values. This information induces a specific metric and the value of the membership function is defined as the normalized distance from the point to the margin of the corresponding crisp set. We show how to build the membership function for the intersubject fuzziness and present a method for interactive refinement of the obtained curves.