关键词:数据聚类;医学图像;图聚类;差分进化 Abstract With the development of medical imaging technology, more and more medical images affect medical personnel on diagnosis. In order to more effective manage and utilize these medical images, researchers pay close attention on data clustering in medical images. While medical images are very complicated in structure and have numerous characteristic, and most medical images are high dimensional. So many common data mining methods which are effective on general database do not really have a good effect. It became the focus to research clustering method for medical images. First in this paper, it analyzes the current domestic and research on medical image mining. Second based on above content, it makes use of a new kind of ROI picked up from image to description medical images' characteristic instead of tradition interest. This method introduces medical knowledge into clustering and hold on to a high precision clustering outcome. The research work and innovations of this paper are as follows: (1)Clustering for medical images based on differential evolution. We combine differential evolution ideology into K-means, and introduce medical knowledge into the algorithm processes. But this algorithm needs not K parameters and is not sensitive for the initial center and noise, but does not affect the accuracy of clustering results.