A FUZZY - GENETIC DECISION SUPPOST SYSTEM FOR MEDICAL DIAGNOSIS

Abstract

The automatic diagnosis of diseases like breast cancer, liver disorder and diabetes are real-world medical problems. A novel approach for diagnosing these diseases is undertaken. This approach based on the evolution of the entire fuzzy inference system for each diagnosis problem, shows that, it is possible to obtain diagnostic systems exhibiting high performance, coupled with interpretability and a confidence measure. The evolved fuzzy systems are able to accurately predict the outcome of a human decision-making process, while providing an understandable explanation of the underlying reasoning. Fuzzy logic provides a formal framework for constructing systems exhibiting both good classification performance and interpretability. Linguistically, fuzzy system represents knowledge in the form of rules, a natural way for explaining decision processes. Optimization of both knowledge base and rule base is critical to the performance of a fuzzy system. Genetic algorithm, a genetically inspire...