Fuzzy logic and inflammatory protein variations

Abstract

This tutorial style paper has been written for physicians and biologists who are not necessarily familiar with fuzzy set theory and biomedical applications. The field is introduced in the framework of medical diagnosis problems and illustrated with an application to inflammatory protein variations. The model is of special interest in the processing of borderline cases, allowing a graded assignment of diagnoses to patients. Relationships between signs and diagnoses are interpreted as labels of fuzzy sets and it is shown how diagnoses can be derived from soft matching processes. In cases of poor diagnostic classification, appropriate weights are introduced, acting on characterizations of signs, in order to decrease their relative influence. As a consequence, when pattern matching is achieved, the final ranking of inflammatory syndromes assigned to a given patient might change to better fit the actual classification.