Electrocardiogram Signal Classification Using Higher-Order Complexity of Hjorth Descriptor

ECG signal is a bio-potential signal generated by the heart muscle that can be used to detect heart abnormalities. Research on the ECG signal classification becomes a topic which is done mostly by researchers. The goal is to find the simplest algorithm, less computation but still has a good performance. In this research, the Higher Order Complexity of Hjorth Descriptor is used to extract the feature of ECG signal. The testing data consists of three types of ECG signal namely Normal Sinus Rhythm (NSR), Atrial Fibrillation (AF) and Congestive Heart Failure (CHF). K-Nearest Neighbor (K-NN) and multilayer perceptron (MLP) is used for classification the feature of the signal from Hjorth Descriptor result. Our propose method produce 94% accuracy using both MLP and K-NN respectively.