Breast cancer is still one of the most common and deadly types of cancer that affect women worldwide, hence improvements in diagnostic techniques are required for early identification and treatment. This review paper discusses the latest developments in the identification and analysis of breast cancers in medical imaging, as well as the comparative effectiveness of different techniques for segmentation and machine learning algorithms. We systematically analyze research that use advanced machine learning techniques including convolutional neural networks (CNNs), support vector machines (SVMs), and deep learning models in addition to more conventional segmentation techniques like thresholding, region-based, and edge detection methods. In comparison to traditional segmentation techniques, the synthesis of findings from the reviewed literature highlights the superiority of machine learning algorithms, especially deep learning techniques, in obtaining improved accuracy, sensitivity, and specificity in breast tumor detection. Moreover, combining segmentation methods with sophisticated machine learning models appears to be a viable way to improve tumor delineation and classification efficiency. In order to improve breast cancer diagnosis and treatment, this study intends to give researchers and medical professionals a thorough assessment of the advantages, disadvantages, and potential future directions of using these technologies.
