A Novel Classification Scheme for Breast Masses Based on Multi-view Information Fusion
The classification of breast masses into benign andmalignant categories plays an important role in the area ofcomputer-aided diagnosis (CAD) of breast cancer. In this paper,one novel scheme based on multi-view information fusion isproposed, in order to improve the accuracy and the robustness of the classification and reduce the false positive rates. Five contourand shape features of the masses are chosen, and two contourfeatures are introduced. In this paper, 304 ROIs (regions ofinterest) that compose 152 multi-view pairs according to thecorresponding breasts are chosen from DDSM dataset anddivided into three groups, according to the characteristics of themasses in each pair. Different groups of ROIs are respectivelyused to check the effectiveness of the classification scheme and the experimental results of three ROIs groups demonstrate that the proposed classification scheme achieves a higher accuracythan those schemes using the individual classifiers in single-view,and the two introduced features could improve the classificationperformance significantly.
Li Sun Lihua Li Weidong Xu Wei Liu Juan Zhang Guoliang Shao
Institute for Biomedical Engineering and Instrumentation Hangzhou Dianzi University Hangzhou, China Department of Radiology Zhejiang Cancer Hospital Hangzhou, China
国际会议
成都
英文
1-4
2010-06-18（万方平台首次上网日期，不代表论文的发表时间）