RETRIEVAL-DRIVEN MICROCALCIFICATION CLASSIFICATION FOR BREAST CANCER DIAGNOSIS

In this paper, a content-based mammogram retrieval system is developed and evaluated to improve the performance of both human and numerical observers in breast cancer diagnosis. We previously developed a machine learning approach for modeling the similarity measure between two lesion mammograms from expert observer studies for mammogram retrieval. In this work we investigate how to use the retrieved similar cases as references to improve a numerical classifier\´s performance. The rationale is that by adaptively incorporating proximity information to the cost function of a classifier, it can help to improve the classification accuracy, thereby leading to an improved "second opinion" to radiologists. Our experiment results on a mammogram database demonstrate that the proposed retrieval-driven approach with an adaptive support vector machine (SVM) could improve the classification performance from 0.78 to 0.82 in terms of the area under the ROC curve