Title: Hybrid technique for the detection of suspicious lesions in digital mammograms

Authors: Laxman Singh; Zainul Abdin Jaffery

Addresses: Department of Electronics & Communication Engineering, Noida Institute of Technology and Management, Greater Noida 201306, India ' Department of Electrical Engineering, Jamia Millia Islamia, Jamia Nagar, New Delhi 110025, India

Abstract: This paper presents an efficient system for the detection of suspicious lesions in mammograms. The proposed detection system consists of three steps. In the first step, an efficient pre-processing technique is developed using Top-Hat morphological filter and NL means filter. In the second step, threshold selection procedure is developed using a combination of Fuzzy C-means (FCM), gradient magnitude (GM), and intensity contrast (IC). Finally, computed threshold is used to extract the suspicious lesions in mammograms. The Free Response Operating Characteristics (FROC) curve is used to assess the performance of the proposed system. Proposed system achieved the sensitivity of 93.8% at the rate of 0.51 false positives per image.

Keywords: mammograms; segmentation; breast cancer; fuzzy C-means; computer-aided diagnosis.

DOI: 10.1504/IJBET.2017.084667

International Journal of Biomedical Engineering and Technology, 2017 Vol.24 No.2, pp.184 - 195

Received: 29 Jan 2016
Accepted: 16 May 2016
Published online: 20 Jun 2017 *

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