Medical Image Retrieval Based on Extraction of Region of Interest

This paper presents some methods of medical image retrieval technique based on extraction of region of interest. First of all, in order to extract the focus region about medical, we propose a middle-value filtering idea combined region growing and mathematical morphology operator. Then based on the idea above, we present two methods of interest extraction. One is the T-Sobel edge detection method that is combined threshold-based segmentation and Sobel operator. The other is based on a method of fuzzy c-means and region growing, which is called FCM-Region. At last, we compare these approaches with the method of retrieval based on medical features and make a detailed analysis in which the result indicates that the method discussed in this paper is much more effective.