A fully automated breast separation For mammographic images

Mammography is currently the most efficient imaging technique employed in radiology for examining breast cancer. Searching for a suitable, flexible and efficient breast profile segmentation method has been shown to be a herculean task in digital mammography. The extraction of the breast profile region is a fundamental pre-processing step in computer- aided detection of breast cancer. Principally, it allows the investigation of abnormalities to be limited to the region of the breast tissue without redundant effect from the image background. In addition, mammograms are very difficult to interpret, mainly of breast cancer. Also, the prevalence of artifacts and noises can interrupt breast cancer detection and reduce the accuracy rate of computer-aided analysis (CAD). Thus, breast profile segmentation of mammogram images is very important since it could reduce the volume of false positives, and also aid radiologists in carrying out comparative analysis between mammograms. The objective of this study is to examine an algorithm of automated breast profile segmentation for mammographic images. The main contribution of proposed algorithm is applying the combined of thresholding technique and morphological preprocessing to segregate background region from the breast profile and remove radiopaque artifacts and labels. To show the validity of our segmentation system, it is extensively tested using over all mammographic images from the MIAS database. The MIAS database comprises 322 images with high intensity rectangular labels. Bright scanning artefacts were found to be present in majority of the database images. All square high intensity labels, apart from three were removed at a rate of 99.06%. The qualitative assessment of experimental results indicates that the method can accurately segment the breast region in a large range of digitised mammograms, covering all density classes.