Title: Robust automatic classification of benign and malignant microcalcification and mass in digital mammography

Authors: J.S. Leena Jasmine; S. Mary Joans

Addresses: Department of Electronics and Communication Engineering, Velammal Engineering College, Ambathur, Redhills Road, Chennai-66, India ' Department of Electronics and Communication Engineering, Velammal Engineering College, Ambathur, Redhills Road, Chennai-66, India

Abstract: Breast cancer is the most dangerous cancer among women and second mortality among them. Mammography is the efficient methodology used in early finding of breast cancer. However, mammograms requires high amount of skill and there is a possibility of radiologist to misunderstand it. Hence, computer aided diagnosis are used for finding the abnormalities in mammograms. Automated classification of mass and microcalcification system is proposed in this work using NSCT and SVM. The classification of abnormalities is achieved by extracting the microcalcification and mass features from the contourlet coefficients of the image and the results are used as an input to the SVM. The proposed automated system classifies the mammogram as normal or abnormal and result is abnormal, then it classifies the abnormal severity as benign or malignant. The evaluation of the proposed system is conceded on MIAS database. The experimentation result shows that the proposed system contributes improved classification rate.

Keywords: mammograms; mass; benign microcalcification; non-subsampled contourlet transform; NSCT; malignant microcalcification; support vector machines; SVM; automatic classification; digital mammography; breast cancer; feature extraction.

DOI: 10.1504/IJBIDM.2016.081869

International Journal of Business Intelligence and Data Mining, 2016 Vol.11 No.3, pp.282 - 298

Received: 10 Sep 2016
Accepted: 10 Oct 2016
Published online: 29 Jan 2017 *

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