Computer-aided Detection Of Mammographic Masses Using Content-based Image Retrieval

Breast cancer is the most common cancer for women in the world. Mammography is currently the most reliable and effective method for detection of breast cancer. Researches indicate that computer-aided detection(CAD) provides an effective second opinion for the diagnosis of breast cancer.A method for computer-aided detection (CAD) of mammographic masses is proposed and a prototype CAD system is presented. The method is based on content-based image retrieval (CBIR). A mammogram database containing 2000 mammographic regions is built in our prototype CBIR-CAD system. Every region of interested (ROI) in the database has known pathology. Specifically, there are 583 ROIs depicting biopsy-proven masses, and the rest 1417 ROIs are normal. Whenever a suspicious ROI is detected in a mammogram by a radiologist with or without the aid of computer, he can submit the ROI as a query to this CBIR-CAD system. As the query results, a serial of similar ROI images together with their known pathology knowledge will be retrieved from the database and displayed in the screen in descending order of their similarities to the query ROI to help the radiologist to make the diagnosis decision. Furthermore, our CBIR-CAD system will output a decision index (DI) for the query ROI to quantitatively indicate the probability that the ROI indeed contains a mass. The DI is calculated by the query matches. In the querying process, 32 features are extracted initially form each ROI as a feature set, then 24 features are selected from the feature set using l-r feature-selecting-method to form a 24-dimension vector. The inverse of the square of Euclidean distance in the 24-dimension feature vector space is applied to measure the similarities between ROIs.The prototype CBIR-CAD system is evaluated based on the leave-one-out sampling scheme and the receiver operating characteristic (ROC) analysis. The experiment results showed that the system can achieve an ROC area index AZ = 0.841 for detection of mammographic masses, which is better than the result obtained by the artificial neural network method developed by the Pittsburg University, USA.CBIR-based CAD is a useful method for computer-aided detection of mammographic masses.