Abstract: Aimed at the problem that the traditional algorithm has poor denoising effect on the interference noise of electrocardiogram (ECG) electromyography, an optimized variational mode decomposition (VMD) combined with improved wavelet threshold denoising algorithm is proposed. The parameters of VMD were optimized by genetic algorithm (GA), and the ECG signal with EMG interference was decomposed into multiple intrinsic mode functions (IMFs). The improved wavelet threshold was used to denoise the IMFs with smaller PCC value. The denoised ECG signal was obtained by reconstructing the every IMFs. The algorithm was compared with other algorithms for ECG signals denoising results with simulated and real EMG interference noise. The results show that the proposed algorithm has small computational complexity and the ECG signals’ useful waveform characteristic can be better maintained after denoising. Moreover, the denoised ECG signals’ signal-to-noise ratio (SNR), mean square error (MSE) and Pearson correlation coefficient (PCC) value are improved to some extent.