EPJ Plus article on Breast cancer: latest improvements in mammography selected for Springer Nature Grand Challenges Programme

A novel technique provides high performance in the analysis of mammographic images

Breast cancer is a disease predominantly affecting females and in the last decades the incidence rate rose. Nowadays, main risk factors, apart from genetic predisposition, include obesity, physical inactivity, hormone replacement therapy during menopause, and alcohol consumption. During the 1980s and 1990s, mammography screening has taken hold detecting many new cases. This technique takes advantage of low energy X-rays to examine breast tissues and early detect masses or microcalcifications, which are cancer’s ‘alarm bells’. Major issues in mammography concern the development of methods allowing a fast and clear interpretation of the collected screening images.

A group of scientists (B. Mughal et al.) reports on the European Physical Journal Plus (EPJ Plus) a new technique to improve the screening images reconstruction in order to achieve high accuracy. The proposed method can be used to remove pectoral muscle from the images which generates noisy features on the detected tumor mass, as well as to segment the breast lesion region providing a larger view of a small section of the image.

The new algorithm was tested on total 513 images taken from the mammography analysis society (MIAS) and from the digital database for screening mammography (DDMS). The algorithm improved the performance of mass segmentation at maintaining the good visual integrity and the overall accuracy obtained by the method on MIAS and DDSM images is 98 percent and 97 percent respectively.

Bushra Mughal, Nazeer Muhammad and Muhammad Sharif (2018), Deviation analysis for texture segmentation of breast lesions in mammographic images, Eur. Phys. J. Plus 133:455, DOI 10.1140/epjp/i2018-12294-4

Reference on the Springer Nature Grand Challanges page

Open calls for papers