Mammogram Classification Using Association Rule Mining

Breast cancer is the primary and the most common disease found among women. It is responsible for rapid growth in mortality rate among all types of cancers in women. Today, mammography the most powerful screening technique is used for early detection of cancer which increases the chance of successful treatment. Screening with mammography can show changes in the breast up to 2-3years before a physician can feel them. But it is not a perfect solution due to abnormalities that are not observable, abnormalities that are misinterpreted and technical problems in the imaging process. Therefore there is a significant need of computer aided detection system which can produce intended results and assist medical staff for accurate diagnosis at an early stage of breast cancer. Much research has been done in the field of mammogram classification during last two decades. With all this effort, there is still no widely used method for mammogram classification because this domain requires high accuracy. With this objective, an attempt is made to build mammogram classification system using association rule mining. In this paper an efficient association rule mining method is proposed to mine hidden relationships and trends in the digital mammograms for accurate classification. Experiments are carried out using MIAS Image Database. The performance of the proposed method is compared with the standard Apriori algorithm. It is found that the proposed method performs better due to reduction in multiple times scanning of database which results in less computation time. Also the mammogram classification system using this method can provide better accuracy up to 90% as compared with the other associative classification techniques. Thus this paper illustrates the use and effectiveness of association rule mining for mammogram classification.