ECG Classification with Modification of Higher-Order Hjorth Descriptors

Abstract

According to ECG signal that refers to a recording of the electrical changes that accompany each cardiac cycle so it can be used to detect and classify heart diseases. In this research, Hjorth Descriptors, which consists of 5 parameters: Activity, Mobility, Complexity, Chaos and Hazard, is used as the estimators for feature extraction. To show the comparative classifications, the Least-Squares (LS), Maximum-Likelihood (ML), Radial Basis Function Network (RBF) and Support Vector Machine (SVM) classifiers were evaluated for their performance in classification. There were three specific types of ECG signal samples, which are Normal Sinus Rhythm (NSR), Atrial Fibrillation (AF) and Congestive Heart Failure (CHF), analyzed and classified. Experiment results show that the alternative Hjorth descriptor could gain more insight of different significance among various types of ECG waveforms representing the heart function in affective condition.