Title: Detection of masses in mammographic images using geometry, Simpson's Diversity Index and SVM

Authors: Andre Pereira Nunes, Aristofanes Correa Silva, Anselmo Cardoso De Paiva

Addresses: Department of Informatics, Federal University of Maranhao – UFMA, Av. dos Portugueses, SN, Campus do Bacanga, 65085-580, Sao Luis, MA, Brazil. ' Department of Electrical Engineering, Federal University of Maranhao – UFMA, Av. dos Portugueses, SN, Campus do Bacanga, 65085-580, Sao Luis, MA, Brazil. ' Applied Computing Group, Federal University of Maranhao, Av. dos Portugueses, SN, Campus do Bacanga, 65085-580, Sao Luis, MA, Brazil

Abstract: This paper presents a computational methodology to detect masses in mammographic images. In the first step, the K-means clustering algorithm and the template-matching technique are used to detect suspicious regions. Next, geometry and texture features of each region are extracted. Texture is described using Simpson|s Diversity Index, which is used in Ecology to measure the biodiversity of an ecosystem. Finally, the information of texture is used by Support Vector Machine (SVM) to classify the suspicious regions into two classes: masses and non-masses. The tests demonstrate that the methodology has 83.94% of accuracy, 83.24% of sensitivity and 84.14% of specificity.

Keywords: mammography; computer-aided detection; K-means clustering; template matching; Simpson|s diversity index; SVM; support vector machines; mammographic images; feature extraction; breast radiography; breast cancer; image processing; breast tissues; abnormal tissues.

DOI: 10.1504/IJSISE.2010.034631

International Journal of Signal and Imaging Systems Engineering, 2010 Vol.3 No.1, pp.40 - 51

Published online: 13 Aug 2010 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article