Fuzzy knowledge-intensive case based classification for the detection of abnormal cardiac beats

Abstract

This paper presents a new automated diagnostic system to classification of electrocardiogram (ECG) cardiac beats. We have developed an intensive-knowledge case based reasoning classifier which uses a distributed case base enriched by partial domain knowledge (rules). An original similarity measures is proposed by combining the sigmoid similarity function with the fuzzy sets to ameliorate the system accuracy in the detection of cardiac arrhythmias. The experiments presented in this work concern the detection of Premature Ventricular Contraction PVC, normal and abnormal cardiac beats from a pattern extracted from the Electronic medical records collected and published by Beth Israel Hospital (MIT-BIH). The achieved results demonstrate the efficiency and the performance of the developed system.