IntroductionThe graphical depiction of the heart beats in the form of electrical signals is known as electrocardiogram (ECG). The abnormal rhythms of the heart beats are termed as cardiac arrhythmias. Some arrhythmias are life threatening. Therefore there is need to identify the heart conditions of cardiac patients. The identification of cardiac arrhythmias in early stage can save the patients from sudden cardiac arrest.A variation in the consecutive cardiac beats is referred to as heart rate variability (HRV). By means of HRV analysis technique cardiac health can usually be computed. An estimation of HRV is recently being adopted as investigation tool for recognition of heart abnormalities in cardiology. AbstractThe study of Heart rate variability is recently gained momentum for an estimation of heart health. This paper suggests a new approach for enhancement of the prediction accuracy of Multi-Layer Perceptrons (MLP) neural network using improved Particle Swarm Optimization (IPSO) technique. The IPSO computes the weights and biases of MLP for the more accurate prediction of the cardiac arrhythmia classes. This study for heart condition prediction involves selection of Three types of heart signals including Left Bundle Branch Block (LBBB), Normal Sinus Rhythm (NSR), Right Bundle Branch Block (RBBB) from MIT-BIH arrhythmia database, formation of heart rate time series, extraction of features from RR interval time series, implementation of training algorithm and prediction of arrhythmia classes. Several experiments on the proposed training method are carried out to superior the convergence ability of MLP. The experimental results gives comparably better evaluation over gradient based Back-Propagation (BP) learning algorithm.