Two‐step evidential fusion approach for accurate breast region segmentation in mammograms. Issue 11 (1st November 2018)

• Record Type: Journal Article
• Authors: Lajili, Rihab; Kalti, Karim; Touil, Asma; Solaiman, Basel; Essoukri Ben Amara, Najoua
• Is Part Of: IET image processing. Volume 12:Issue 11(2018)
• Page Start: 1972
• Page End: 1982
• Publication Date: 2018-11-01
• Subjects: mammography -- medical image processing -- image segmentation -- image fusion -- iterative methods; two‐step evidential fusion approach -- accurate breast region segmentation -- mammograms -- breast skin line -- breast organ compression -- image acquisition process -- breast region delimitation -- complementary segmentation techniques -- grey‐scale intensities -- local‐homogeneity domains -- knowledge source -- belief functions formalism -- iterative process -- ambiguity removal -- decision quality improvement; Image processing -- Periodicals; 621.36705
• DOI: 10.1049/iet-ipr.2018.5325
• Languages: English
• ISSNs: 1751-9659

Abstract

Abstract : In mammograms, the breast skin line often appears ambiguous and poorly defined. This is mainly due to the breast organ compression during the image acquisition process along with the inherent low density of the tissue in that area. The accurate delimitation of the breast region becomes a challenging task to conventional segmentation techniques. In this study, the authors propose a new segmentation approach allowing to overcome this challenge. This approach is based on the application of two complementary segmentation techniques exploring each, respectively, the grey‐scale intensities and the local‐homogeneity domains. The knowledge resulting from each segmentation technique is considered as a knowledge source and is modelled using the belief functions formalism. The two considered knowledge sources are then fused using an iterative process. The obtained results show the efficiency of the proposed evidential approach especially in terms of ambiguity removal and decision quality improvement for accurate breast border delimitation (which is often under‐segmented and assimilated to the background by most of the existing segmentation techniques).