De-noisng of ECG based on EMD improved-thresholding and mathematical morphology operation

Suppressing the noise of ECG effectively is a prerequisite for the intelligent analysis of ECG. This paper proposed a new ECG denoising method which combined empirical mode decomposition (EMD) with mathematical morphology operation (MMO). First of all, apply the EMD to decompose the noisy ECG into a series of intrinsic mode functions (IMF); then, estimate the intensity of noise for each IMF by MMO, and do improved-thresholding on each IMF based on 3σ criterion; finally, reconstruct the signal with the processed IMF to get the de-noised ECG. The experimental results show that the proposed method has better performance than that of single method.