The paper presents a genetic algorithm based on wavelet threshold medical image denoising method. Theoretically analyses the principle of wavelet threshold denoising, the improved threshold function is proposed. Using genetic algorithm combined with clustering histogram of the image information are used to get the optimal threshold, the calculation without a priori information, such as noise variance and improve search efficiency. Experimental results show that with the ordinary wavelet threshold demising method, this method can improve the visual effect of medical image after denoising, improve the peak signal to noise ratio.