A Hybrid Fuzzy-Neural Expert System for Diagnosis

Fuzzy Logic, a neural network and an expert system are combined to build a hybrid diagnosis system. With this system we introduce a new approach to the acquisition of knowledge bases. Our system consists of a fuzzy expert system with a dual source knowledge base. Two sets of rules are acquired, inductively from given examples and deductively formulated by a physician. A fuzzy neural network serves to learn from sample data and allows to extract fuzzy rules for the knowledge base. The diagnosis of electroencephalograms by interpretation of graphoelements serves as visualization for our approach. Preliminary results demonstrate the promising possibilities offered by our method. 1 Introduction Repetitively applied cognitive tasks of recognizing and evaluating certain phenomena, called diagnostic tasks, are among the main applications for Artificial Intelligence (AI). As there exists a vast variety of such diagnostic tasks in medicine, it has always belonged to the spectrum of potential users of Artificial Intelligence. Most popular among AI methods in medicine are knowledge based systems [Buchanan and Shortliffe, 1985], modeling the diagnostic behaviour of experts. A variety of such expert systems is being used in everyday p ractice of physicians since Shortliffe introduced MYCIN Shortliffe, 1976], an expert system designed to diagnose infections of the human blood. One of the greatest difficulties in designing a convenient expert system is acquiring the knowledge base. We introduce a new approach where a dual source knowledge base is generated by de-ductive and inductive learning. Neural networks have also made their way into diagnosis. They are able to learn relationships between data sets by simply having sample data represented to their input and output layers. In the field of pattern recognition in medical data, neural network based approaches have led to quite remarkable results, for example in processing MRI pictures [Hall et a/., 1992] or EEG traces [Mamelak et a/., 1991; Jando et a/., 1993]. For the task of acquiring knowledge bases, which is a part of our hybrid approach, neural networks have been proposed recently [Thrun and Mitchell, 1993]. Fuzzy logic [Zadeh, 1965] also makes its appearance in medicine, dealing with the uncertainty of verbal expressions [Kuncheva, 1991; Nishimura et a/., 199l]. Terms like many, few or probably are hard to model with conventional logic. The linguistic variables offered by fuzzy representations allow pseudo-verbal descriptions close to natural human expressions. All of the above methods bear advantages as well as disadvantages as will …