A Combined Approach to Medical Image Segmentation Using Anisotropic Diffusion and Wavelet Packet Decomposition

In this paper, a new algorithm to extraction of anatomical contour in medical image has been presented, which combining anisotropic diffusion and wavelet packet decomposition. First, we explored a new anisotropic diffusion model combined with median filter, which was used for both image smoothing and noise removal, as well as edge enhancement. Then, a self-adaptable edge detection algorithm based on wavelet packet decomposition was applied to detect edge for image. Finally, Examples are presented to demonstrate the efficiency of the approach on spinal CT/MRI images and the results are encouraging.