An adaptive level dependent wavelet thresholding for ECG denoising

Abstract

This paper describes the research carried out to eliminate the noise found in ECG signal and cardiac rhythm. For this, ECG signals were collected carefully from BIOPAC data acquisition system and MIT-BIH database. MIT-BIH noise stress test database was used for generating realistic noises. In addition, to get a better denoised ECG, Symlet wavelet was chosen because its scaling function is closely related to the shape of ECG. For denoising ECG signal, a novel modified S-median thresholding technique is proposed and evaluated in this paper. The optimal Symlet wavelet of order 6 and decomposition level of 8 are attained for modified S-median thresholding technique. The evaluation results showed that the proposed system performed better than S-median and other existing techniques in the time domain. The frequency domain analysis also showed the preservation of important phenomena of ECG. The scalogram difference of 0.004% indicates the well preservation of time–frequency information.