The Optimal De-noising Algorithm for ECG Using Stationary Wavelet Transform

The artifacts of ECG signals include baseline wander (BW), muscle (EMG) artifact, electrode motion artifact and power line interference. In order to get the optimal and robust de-noising algorithm among the generally used de-noising methods based on stationary wavelet transform (SWT), we adjust the signal-to-noise ratio (SNR) of the noisy signal from 1 db to 10 db, and evaluate the results by means of SNR and visual inspection, then conclude using Symlet4, decomposition at level 5, and hard shrinkage function with empirical Bayesian (EBayes) threshold can get consistently superior de-noising performance. In addition, test the proposed algorithm using MIT-BIH noise stress database, the results demonstrate that the proposed method improves the SNR and preserves the waveform, which can be used for clinic analysis.