Mayank Yadu, Jagvir Verma
Abstract: One of the scientific tests performed to diagnose Heart Diseases is through Electrocardiogram signals, which shows an electrical activity of the heart in terms of the waves. To diagnose Heart Diseases based ON ECG signal, a medical doctor obtained the features like amplitude of the waves QRS-complex, P-wave and T-wave and the time interval between the waves called R-R interval, P-R interval, S-T interval and Q-T interval. Since an ECG signals may be of different lengths and as being a non-stationary signal, the irregularity may not be periodic instead of showing up at any interval of the signal, a physician has to analyse the signals completely which is a time consuming process. Therefore, in the present study, an algorithm is developed to pre-process and to automatically extract the features from ECG signal based on Discrete Wavelet Transform (DWT) and De-noising factor. The developed algorithm initially performs pre-processing of a signal in order to remove Baseline Drift (De-trending) and to remove noise (De-noising) from the signal and then it uses the pre-processed signal for feature extraction from the ECG signal automatically. By using this algorithm the accuracy of the analysis can improved and the analysis time of an ECG signal can be reduced.
Keywords: ECG, QRS-complex-wave, T-wave, R-R interval, P-R interval, S-T interval and Q-T interval, DWT, Baseline Drift, Denoising