Comparative Analysis of Mammography Image Segmentation Strategies

Abstract

Breast cancer is a serious medical problem that affects women all over the world, and it is one of the most well-known tumors that kill women. The specialists of Breast cancer Prefer to use imaging methods such as a mammography to speed up recovery and reduce the risk of breast cancer. An ROI describe the tumor will be retrieved from the image that is entered to detect a malignant tumor. One of the basic techniques used to classify breast cancer is segmentation. Segmentation may be difficult in the presence of noise, blurring or low contrast. Pre-processing aids in the removal of extraneous data from a picture or the enhancement of image contrast in the early stages. Classification is greatly influenced by segmentation. Recent research have presented automatic and semi-automated segmentation algorithms for extracting the region of interest (ROI), lesions, and masses to check for breast cancer. In this study provides high-level overview of approaches of segmentation, with a focus on mammography images from current research. The datasets that were available were discussed as well as the problems encountered during the segmentation operation for the identification of breast cancer.