Feature extraction of ECG signal based on wavelet transform for arrhythmia detection

Electrocardiogram (ECG) signal provides valuable information about the functional aspects of the cardiovascular system. Cardiac arrhythmia is a condition of abnormal electrical activity in the heart which could be analyzed by ECG signal analysis. The morphological features can be extracted from the signal and be used for the classification of heart beats according to different arrhythmias. The objective of this work is to detect the cardiac arrhythmia automatically from ECG signal based on the detection of QRS complex and R-peak based on wavelet transform. Data were obtained from the MIT-BIH arrhythmia database. The wavelet transform function is used to detect the peaks and QRS complex in the ECG signal to identify the abnormality in the recorded signal. The performance of the DWT based QRS and peak detectors is outperformed to detect the peak values and the on and off sets of different peaks. The compared result shows that the morphological values are better in both amplitude and QRS duration on normal ECG signal.