MR brain image segmentation based on wavelet transform and SOM neural network

Magnetic resonance (MR) brain image has been accepted as the reference image in the clinical research. The goal of MR brain image segmentation is to accurately identify the principal tissue structures in the image volumes. In this paper, the segmentation algorithm based on SOM (self-organizing map) neural network with compression pre-processing by wavelet transform is presented. The compression idea origins from image pyramid structure theory, which can enhance the representation of later image feature extraction without affecting brain tissue structure information. Compared with the traditional individual SOM network method, the hybrid method can improve network training quality by applying statistical intensity information of the compression image pixels as network input vectors. Simulated MR brain images with different noise levels and intensity inhomogeneities are segmented to demonstrate the superiority of the proposed method compared to the traditional technique.