Objective Application of Fuzzy C-means clustering algorithm for image segmentation of magnetic resonance imaging in mice, according to the imaging characteristics of different tissues and organs, segmentation to different organizations. Comparison the image segmentation algorithm with different image features processing. Methods 1. Obtain the mice MRI, scanning mice with the 7.0T MR(Bruker Biospec70cm/ 20cm), the basic parameters:TR / TE = 4000 / 45ms, thickness:0.5mm, the image matrix 256 * 256. 2. Image pre-processing ,frist making a unevenly field correction by SPM5,and then dividing the image into three groups (A, B, C) ,based on different pre-processing methods : group A set of images do mean filtering(module size 3 * 3); group B do the standard differential image processing(module size 3 * 3); group C set of images not be processed . 3. Construction the image segmentation program based on fuzzy C - means clustering algorithm , the process is as follows: pre-processing and mapping the input image to the 256 steps gray scale image, drawing the region of interest (ROI), calculating the mean values of the ROI ares as a initial values of the clustering algorithm, Setting iterative loop condition, computing the cluster center membership and updated iteration, showing segmentation results after each iteration of clustering and renderings completed. 4. In the Matlab platform programming the segmentation algorithm, setting the clustering initial values,the number of clusters and the clustering algorithm termination condition. Respectively segmentation the three groups(A,B,C)of image. Results A,B,C three groups after segmentation the image can get a good result. The result can accurately distinguish between background noise and the region of mouse tissue, can split the cortex of mice, the lateral ventricles, and other soft tissue and skull organizational structure. The result of A group with less noise signal can split the complete anatomy.The result of group B shows a strong gradient boundary signals can split out more anatomical information,but regional poor. The result of group C with anatomical structures, and the mice segmentation shows between group A and group B . Conclusion Fuzzy C-means clustering algorithm can segment the mice MRI while. In mice MRI image processing,it plays an important role.