Najmeh Ghadiri Hedeshi - Faculty of Electrical and Computer Engineering, Tarbiat Modares University
Mohammad Saniee Abadeh - Assistant Professor, Faculty of Electrical and Computer Engineering, Tarbiat Modares University

It is evident that usage of data mining methods in disease diagnosis has been increasing gradually. In this paper,diagnosis of Coronary Artery Disease, which is one of the most well-known diseases that cause heart failure, was conducted with such a data mining system. Many researchers have attempted to develop a medical expert system to increase the ability of physicians in detecting this disease. This paper proposes a new ensemble PSO-based approach to extract a set of rules for diagnosis of coronary artery disease. The new presented boosting mechanism considers the cooperation between generated fuzzy if-then rules using the PSO metaheuristic. We have evaluated our new classification approach using the well-known Cleveland data set. Results indicate that the proposed learning method can detect the coronary artery disease with an acceptable accuracy. In addition, the extractedfuzzy rules have significant interpretability either.

boosting algorithm, BRAMS concept, classification, coronary artery disease, fuzzy logic, Particle SwarmOptimization