Diagnosis Technology Research Of Mammographic Masses Based On Content-based Image Retrieval

Breast cancer is one of the most dangerous diseases to the women. The mammogram is the most reliable clinical method to detect breast diseases. But the mammograms masses are usually difficult to detect as they often superimpose on dense structured background. Thence, computer aided diagnosis system based digital image processing has gradually become focus of the medical image.A method for computer-aided diagnosis (CAD) of mammographic masses is proposed and a prototype CAD system is presented in the dissertation. The method is based on content-based image retrieval (CBIR). The main contents are as follows:1. A mammogram database is built and contains 835 mammographic regions. Every region of interested (ROI) in the database has known pathology ?benign and malignant. It is the base of data according to ROI queries.2. Due to the difficult problem that how to segment the mass region of the queried ROI, an interactive mass segmentation method is given. The size of mass by user provided is used. The improved growing algorithm is adopted to segment the suspicious region of mass, and evaluated by the method of overlapping percentage. For the effective feature selection problem, 32 features are extracted initially form each ROI as a feature set, then 22 features are selected from the feature set using l-r feature selecting method to form a 22 dimension vector.3. To the accuracy of similarity matching of CBIR as the standard, the scheme proposed a method of similarity measurement that combined k-nearest neighbor (KNN) algorithm and the mutual information (MI). A k-nearest neighbor (KNN) algorithm is used to retrieve similar images from database, and further calculate the mutual information (MI) between the queried image and the images which are in the retrieval results, so as to improve the retrieval performance.4. With the purpose of providing available decision-making information of diagnostic aids, four calculating methods of decision index are compared and analyzed, and a stable and available method of decision indes is adopted. The prototype system is evaluated based on the receiver operating characteristic (ROC) analysis. The Experiment results show that the method is better than method of using KNN only, and improve the accuracy of diagnosis effectively.This system is a useful method for computer-aided diagnosis of mammographic masses.