Application of wavelet threshold method based on optimized VMD to ECG denoising

Abstract

Aiming at the problems of weak ECG signal (ECG), low frequency range and incomplete noise removal, a wavelet threshold denoising algorithm based on optimized VMD (Variational Modal Decomposition) is proposed. This method first reads the ECG signal, and uses genetic algorithm (GA) to optimize the two parameters of the VMD decomposition level and the penalty factor in the parameter search space. After the parameters are optimized, the ECG signal is decomposed by VMD, and a series of decompositions are obtained. Modal component (u), judge the main component of each modal component according to the correlation coefficient of each modal component and the noisy ECG signal, improve the wavelet threshold denoising of the noise-dominated modal component, and denoise the denoised modal The component and the modal component dominated by the ECG signal are inversely transformed and reconstructed to obtain the denoised ECG signal. Use the real ECG signal in the MIT-BIH database to conduct experiments, and compare the wavelet threshold denoising method based on optimized VMD with the wavelet threshold method alone and the wavelet threshold method based on VMD. The results show that the method has better denoising effect. good.