In order to solve the problem of several common methods in medical ultrasound image processing which includes Poor retention of detailed information and insignificant denoising effect, therefore a new method of ultrasonic image denoising combining two-dimensional variational mode decomposition (abbreviated as 2D-VDM) and anisotropic diffusion (abbreviated as PM) is proposed. This method firstly decomposes the image into a series of modal component (IMF) images through two-dimensional variational mode decomposition (2D-VDM),and then uses the peak signal-to-noise ratio and the normalized mean square error to filter out the effective modal components, finally, the effective modal components are subjected to anisotropic diffusion (PM) filter processing and reconstruct the processed effective components to remove image noise.The evaluation of image quality indicators from peak signal-to-noise ratio and root mean square error shows that this method is superior to other commonly used methods in removing noise and protecting detailed information in the image.