Normal mammogram classification based on a support vector machine utilizing crossed distribution features

Author

Chiracharit, W. ; Sun, Y. ; Kumhom, P. ; Chamnongthai, K. ; Babbs, C. ; Delp, E.J.

Author_Institution

Dept. of Electron. & Telecom. Eng., King Mongkut´´s Univ. of Technol., Bangkok, Thailand

Volume

1

fYear

2004

fDate

1-5 Sept. 2004

Firstpage

1581

Lastpage

1584

Abstract

Automatic classification of normal mammograms, which constitute a majority of screening mammograms, is a new approach to computer-aided diagnosis of breast cancer. This approach may be limited, however, by non-separable "crossed" distributions of features that are extracted from digitized mammograms. This work presents a method of mapping such non-separable input features into a new set of separable features that can be utilized, together with ordinary "uncrossed" features, by a support vector machine (SVM) classifier. The results of the proposed scheme show improved performance with 80% sensitivity and 95% specificity.

Keywords

biological organs; cancer; feature extraction; image classification; mammography; medical image processing; support vector machines; breast cancer; computer-aided diagnosis; crossed distribution features; digitized mammograms; feature extraction; mammogram classification; nonseparable input features; screening mammograms; support vector machine; Aging; Biomedical engineering; Biomedical imaging; Breast cancer; Computer aided diagnosis; Data mining; Mammography; Support vector machine classification; Support vector machines; Training data; Breast cancer; computer-aided diagnosis (CAD); mammogram; support vector machine (SVM);

fLanguage

English

Publisher

ieee

Conference_Titel

Engineering in Medicine and Biology Society, 2004. IEMBS '04. 26th Annual International Conference of the IEEE

Conference_Location

San Francisco, CA

Print_ISBN

0-7803-8439-3

Type

conf

DOI

10.1109/IEMBS.2004.1403481

Filename

1403481