Cardiac arrhythmias classification using artificial metaplasticity algorithm

Abstract

Electrocardiogram (ECG) arrhythmias such as ventricular and atrial arrhythmias are one of the common causes of death. These abnormalities of heart activity may cause immediate death or damage to the heart. If the abnormal symptoms can be detected and diagnosed early, time is saved to prevent the occurrence of heart attack. Therefore, it is necessary to have an effective method for early detection and early treatment. We propose, in this paper, an intelligent method to accurately classify the heartbeat of ECG signals through the Artificial Metaplasticity Multilayer Perceptron (AMMLP). The MIT-BIH database is used to classify arrhythmias into three different types: Premature Ventricular Contraction (PVC), Right Bundle Branch Block (RBBB) and Left Bundle Branch Block (LBBB); normal ECG signals are also used in the study. The obtained AMMLP classification accuracy of 98.25% is an excellent result compared to the classical MLP and recent classification techniques applied to the same database.