Heart Rhythm Abnormality Detection and Classification using Machine Learning Technique

Abstract

Electrocardiogram (ECG) plays important role in detection and classification of cardiac irregularities. This research presents the approach for classification of heartbeat irregularity. Three different signals Cardiac Arrythmia (ARR), Normal Sinus Rhythm (NSR) and Congestive Heart Failure (CHF) are considered for research. A total of 162 records are considered for research. Then data collected is to be divided into two sets-training set and training set. Training set comprises of 70 percent of data and testing set comprises of remaining 30 percent. The paper mainly follows four stages, in stage 1 Arrhythmia signals and Non- Arrhythmia signals are collected from MIT- BIH database for further study. In stage 2 the collected Cardiac Arrhythmia (ARR), Normal Sinus Rhythm (NSR) and Congestive Heart Failure (CHF) signals are prepossessed. In stage 3 features are extracted from pre-possessed signals using Discrete Wavelet Transform (DWT) and all the features are concatenated into a single feature vector. In stage4 the extracted features are given to Support Vector Machine (SVM) classifier for classification of the model and the parameters such as precision, recall and F1 score are calculated. The accuracy obtained is 95.92 percent.