An efficient pixel value based mapping scheme to delineate pectoral muscle from mammograms

Saltanat, N., Hossain, Alamgir and Alam, Muhammad (2010) An efficient pixel value based mapping scheme to delineate pectoral muscle from mammograms. In: IEEE Fifth International Conference on Bio-Inspired Computing: Theories and Applications, 23-26 September 2010, Changsha, China.

Full text not available from this repository. (Request a copy)

Abstract

Mammograms are X-ray images which are used in breast cancer detection. In Computer Aided Detection of breast cancer from digital mammogram, elimination of pectoral muscle is a very important and challenging issue. This is because of the fact that pectoral muscle in mediolateral oblique (MLO) mammogram images has common photographic properties with suspicious mass and micro-calcification. Presence of pectoral muscle gives false positive result in automated breast cancer detection. In this paper a novel and efficient method using pixel value mapping is proposed to delineate pectoral muscle region accurately. The proposed method is capable of segmenting pectoral muscle of a broad range of size, shape and position. This algorithm is found to be robust not only to large variations of size, shape and positions of pectoral muscle, but also to any kind of artifacts like medical tags. The algorithm has been applied to 322 images of Mammographic Image Analysis Society (MIAS) database. The segmented results were then evaluated by two expert radiologists, who rated 84% and 94% of the segmentations to be accurate respectively. This algorithm is found to be robust not only to large variations of size, shape and positions of pectoral muscle, but also to any kind of artifacts like medical tags

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	automatic segmentation , pixel value mapping, region grow method
Subjects:	B800 Medical Technology G900 Others in Mathematical and Computing Sciences
Department:	Faculties > Engineering and Environment > Computer and Information Sciences
Depositing User:	EPrint Services
Date Deposited:	18 Aug 2011 08:43
Last Modified:	31 Jul 2021 08:38
URI:	http://nrl.northumbria.ac.uk/id/eprint/193

Downloads