22 - Classification of Breast Lesions from Mammograms

Abstract

The effectiveness of breast cancer treatment depends strongly on cancer progression. Compared to advanced cancers, patients diagnosed with minimal cancers have significantly better 20-year survival rate. State-of-the-art mammography is the most effective imaging modality in breast cancer detection, particularly in finding nonpalpable small cancers. Large scale, randomized controlled clinical trials have shown that annual screening of asymptomatic women reduces breast cancer mortality by 20–30%. The results of this study have led to the recommendation and systematic implementation of screening for breast cancer in the United States, Canada, and many European countries. Results of performance studies showing the effect of computer classification on radiologists' performance are also encouraging. These results indicate that radiologists can outperform themselves when they use computer classification as an aid, and improve their diagnoses in two ways. With the computer aid, radiologists can potentially send fewer women without breast cancer to biopsy and at the same time correctly diagnose more cancers. With the aid of computer classification, it is possible to reduce the physical and psychological trauma in patients who have benign breast lesions by avoiding unnecessary biopsies without compromising the effectiveness of mammography screening for breast cancer. As a result, the cost associated with mammography screening could be reduced. However, the computer classification must also be further developed and tested in larger studies before it can be used clinically, and before its potential in reducing the morbidity and mortality of breast cancer can be realized in clinical medicine.