Cardiac Arrhythmia Classification of ECG Signal Using Morphology and Heart Beat Rate

The electrocardiogram (ECG) signal represents the electrical activity of the heart. The most characteristic wave of the ECG is the QRS complex, which records the polarization phenomenon of the ventricles. The objective of the work is to detect and classify cardiac arrhythmias of patients from the ECG signal. The detection of the QRS complex is gaining momentum and many algorithms have been reported in the literature for R-peak detection. To detect the R-peak, a modified Pan-Tompkins [7] approach has been implemented in this work. Further, a Linear Discriminate (LD) function, Scaled conjugate gradient method, and Naive Bayes Classifier approaches are used in this work for classification of cardiac arrhythmias. Classification consists of four features such as R-R interval, QRS interval, QRS morphology and T-wave morphology by extracting from each cardiac cycle. The above mentioned algorithms are implemented in a MATLAB environment with test data obtained from the MIT BIH Arrhythmia database. The performance of the three classifiers is verified using the test data.