Performance Analysis of ANN and SVM in ECG Based Arrhythmia Identification

Abstract

This paper presents a performance analysis of Artificial Neural Network (ANN) and Support Vector Machine (SVM) algorithms in arrhythmia identification task based on ECG signals. Six features are used for both algorithms: short signal 1-D wavelet energy (SS-WVE), short signal continuous wavelet transform mean (SS-CWTM), heart rate (HR), R-peaks root mean square (R-RMS), RR-peaks variance (RR-VAR) and QRS-complex standard deviation (QRS-SD). The identification methods use the MIT-BIH Dataset (Massachusetts Institute of Technology-Beth Israel Hospital) for training, validation and test phases. In this work, preliminary results shown that the classification obtained using SVM is marginally better than the one obtained with the ANN classifier for the same classification task (i.e. arrhythmia pattern identification).