Wavelet denoising of the electrocardiogram signal based on the corrupted noise estimation

Abstract

We present in this paper an algorithm of filtering the noisy real ECG signal. The classical wavelet denoising process, based on the Donoho et al. algorithm, at the 4 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">th</sup> level, appears clearly the P and T waves whereas the R waves undergo considerable distortion. This is due to the interference of the WGN and the free noise ECG detail sequences at level 4. To overcome this drawback, our key idea is to estimate the corrupted WGN and consequently remove the noise interfering R waves at the 4 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">th</sup> level detail sequence. Our denoising algorithm was applied to a set of the MIT-BIH arrhythmia database ECG records corrupted with a 0 dB WGN which provided an output SNR of around 6 dB and an MSE value of around 0.0011. A comparative analysis using the low pass Butterworth filter and the 4 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">th</sup> level classical wavelet denoising provides the output SNR values of around 3 dB and MSE value of around 0.0018; which demonstrates the superior performance of our proposed denoising algorithm