Title: Mammogram mass classification using various geometric shape and margin features for early detection of breast cancer

Authors: B. Surendiran; A. Vadivel

Addresses: Department of Computer Science and Engineering, Kumaraguru College of Technology, Coimbatore – 641049, Tamilnadu, India. ' Department of Computer Applications, National Institute of Technology, Tiruchirappalli, Tamilnadu – 620015, India

Abstract: This paper focuses on an approach for characterising the mammogram masses using various geometric shape and margin features. According to BIRADS system, benign and malignant masses can be differentiated using its shape, size and density features, which is how radiologist visualise the mammograms. According to BIRADS, benign masses are round, oval, lobular in shape and malignant masses are lobular or irregular in shape. Various 17 geometrical shape and margin features are introduced to characterise the morphology of masses, as there is no single measure to differentiate various shapes. Experiments have been conducted on 1553 DDSM database mammograms and classified using CART classifier. Experimental results indicate that CART can classify masses effectively and generates simple rules, which can be easily implemented in any system using if..then..else statements. Experimental results are found to be encouraging. The results demonstrate the effectiveness of CART classifier for classifying masses as benign, malignant and normal.

Keywords: digital mammograms; geometrical shape features; margin features; benign masses; malignant masses; normal masses; classification; regression tree; CART; breast imaging; reporting and data system; BI-RADS^TM categories; computer aided diagnosis; radiology; early detection; breast cancer; early diagnosis; tumours.

DOI: 10.1504/IJMEI.2012.045302

International Journal of Medical Engineering and Informatics, 2012 Vol.4 No.1, pp.36 - 54

Published online: 11 Aug 2014 *

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