Wavelet Based ECG Denoising Using Signal-Noise Residue Method

Noise removal from ECG signal is a classic problem and has been undertaken using different approaches including wavelet based techniques, adaptive filtering, neural networks and model based filtering. Wavelet transform offers obvious advantages over the other approaches in the analysis of non stationary signals such as the ECG owing to its multi-scale decomposition of the signal. Our proposed Signal-Noise Residue method is a novel wavelet based ECG denoising algorithm which offers enhanced performance compared to standard wavelet based techniques. The algorithm was tested on the MIT/BIH Arrhythmia database (MITDB) and the National Institute of Heart Diseases, Pakistan (NIHD) ECG database giving excellent results.