Cardiac arrhythmia detection using linear and non-linear features of HRV signal

Abstract

Earlier detection of Cardiac arrhythmias from long term ECG recording is one of the complex problems in signal processing. In this paper, we proposed an effective algorithm to detect and classify the cardiac abnormalities. By extracting different features in time domain, frequency domain and nonlinear features from heart rate variability (HRV) signals, the algorithm can differentiate between the types of arrhythmias. The features extracted from HRV signal are used to train and test the Support Vector Machine (SVM) classifier to classify Normal Beat, Premature Atrial Contraction (PAC), Right Bundle Branch Block (RBBB), and Paced Beat. The ECG signal is downloaded from MIT-BIH database. Training and testing of classification algorithm yields overall accuracy of 90.26%.