A Texture Features Based Medical Image Retrieval System for Breast Cancer

This paper designed and implemented a content-based medical image retrieval (CBMIR) system for mammograms images. The main functions of this system includes query module, feature extraction module, matching module and display module. The system uses the Euclidean distance as an image similarity measure, using the gray level co-occurrence matrix texture and Tamura texture feature extraction. Experiments are performed with 85 mass images from DDSM dataset, an average precision of 69% is achieved.