A fuzzy approach to discriminating heartbeat types and detecting arrhythmia

Traditional ECG inspection procedures can be very complicated and need professional assistance. When minor abnormality occurs the subject´s heartbeat often returns to normal before he/she takes a detailed examination. Thus, the cause of abnormality remains ignored. In this study, we propose an effective heartbeat monitoring ECG real-time detection system for homecare service, which uses the ECG sensors and a wireless sensor network technology to detect the subject´s heartbeats and their variations. In addition, the MIT-BIH database is used to analyze arrhythmia. Based on the selected features an arrhythmia detection module is devised to detect the arrhythmia that is determined by the proposed fuzzy rules. Experimental results show that the proposed system is able to discriminate normal and arrhythmia heartbeats. For NORM, LBBB, RBBB, VPC, APC and PB the discriminant accuracy rates are 97.5%, 87.5%, 92.5%, 100%, 95% and 100%, respectively.