A weighted fuzzy reasoning algorithm for medical diagnosis

Decision Support SystemsPublished 1 January 1994

Shyi‐Ming Chen

DOI10.1016/0167-9236(94)90063-9

Source URL

Citations131

SJR quartileQ1

SJR score2.37

SNIP2.59

TL;DR

A weighted fuzzy reasoning algorithm for handling medical diagnostic problems, where fuzzy set theory and fuzzy production rules are used for knowledge representation, which can perform fuzzy matching between the patient's symptom manifestations and the antecedent portions of fuzzyproduction rules to determine the presence of diseases.

Abstract

This paper presents a weighted fuzzy reasoning algorithm for handling medical diagnostic problems, where fuzzy set theory and fuzzy production rules are used for knowledge representation. The algorithm can perform fuzzy matching between the patient's symptom manifestations and the antecedent portions of fuzzy production rules to determine the presence of diseases, where the result is interpreted as a certainty level indicating the degree of certainty of the presence of the disease. Because the algorithm allows each symptom in medical diagnosis to have a different degree of importance, it is more flexible than the ones we presented in [3] and [4]. The algorithm can be executed very efficiently. If the knowledge base contains n fuzzy production rules and there are p symptoms, then the time complexity of the algorithm is O(np).

Keywords

Computer Science

Information and ControlFuzzy sets

66,132 Citations1965L. A. Zadeh

Fuzzy Sets and SystemsThe role of fuzzy logic in the management of uncertainty in expert systems

1,282 Citations1983Lotfi A. Zadeh

F fuzzy logic is suggested, which is the logic underlying approximate or, equivalently, fuzzy reasoning, which leads to various basic syllogisms which may be used as rules of combination of evidence in expert systems.

IEEE Transactions on Knowledge and Data EngineeringKnowledge representation using fuzzy Petri nets

540 Citations1990Shyi‐Ming Chen, Jyh-Sheng Ke +1 more

A fuzzy Petri net model (FPN) is presented to represent the fuzzy production rule of a rule-based system in which a fuzzy productionrule describes the fuzzy relation between two propositions and an efficient algorithm is proposed to perform fuzzy reasoning automatically.

IEEE Transactions on Systems Man and CyberneticsFuzzy Set Theory in Medical Diagnosis

424 Citations1986Klaus-Peter Adlassnig

Fuzzy set theory has a number of properties that make it suitable for formalizing the uncertain information upon which medical diagnosis and treatment is usually based, and trials performed with the medical expert system CADIAG-2 suggest that it might be a suitable basis for the development of a computerized diagnosis system.

IEEE Transactions on Systems Man and CyberneticsA new approach to handling fuzzy decision-making problems

264 Citations1988Shyi‐Ming Chen

Techniques for handling fuzzy decision-making problems are presented in which fuzzy production rules and fuzzy set theory are used for knowledge representation and the maximum fuzzy cover generation techniques are described in detail.

ComputerFuzzy concepts in expert systems

142 Citations1988Kwong‐Sak Leung, W. Lam

The authors present a comprehensive expert-system building tool, called System Z-II, that can deal with exact, fuzzy, and combined reasoning, allowing fuzzy and normal terms to be freely mixed in the rules and facts of an expert system.

Expert SystemsApplications of a novel fuzzy expert system shell

27 Citations1989Alexander K. C. Leung, Wing Shing Wong +1 more

It has been demonstrated that the introduction of fuzzy concepts into expert systesms is not feasible but highly desirable and the additional power and advantages over traditional expert systems are illustrated.

Cybernetics & SystemsAN EFFICIENT ALGORITHM TO HANDLE MEDICAL DIAGNOSTIC PROBLEMS

9 Citations1990Shyi‐Ming Chen, Jyh-Sheng Ke +1 more

A new algorithm is proposed to handle medical diagnostic problems, where the fuzzy production rules are used for knowledge representation, and the time complexity of the algorithm is O(np).