Biomedical Signal Processing: ECG Signal Analysis Using Machine Learning in MATLAB

Abstract

An arrhythmia is an abnormality in the heart rhythm or heartbeat pattern. ECG beats can be classified into different arrhythmias beat types (bigeminy, trigeminy, ventricular tachycardia (VT)). Early and accurate detection of arrhythmia types is important in detecting heart diseases and choosing appropriate treatment for a patient. We have used MIT-BIH arrhythmia database for data collection and prepared different data sets. Features, such as amplitude, RR interval, heart rate (Speed), gender, age, are used for the analysis. In classification learner application, the extracted features are used as inputs to different classifiers: support vector machines (SVM) and Naïve Bayes. Some other techniques that also have been employed for arrhythmia classification are decision trees and ensemble learning. Results show high classification accuracy of over 99.3% with either of these classifiers. The performance comparison of these classifiers is carried out using accuracy. Each classifier can show the confusion matrix, which summarizes the accuracy for each true label class.