Segmentation and Classification for Breast Cancer Ultrasound Images Using Deep Learning Techniques: A Review

Abstract

Deep Learning (DL) has rapidly become a methodology of choice for analyzing medical images and increasingly attracts researchers’ attention in the medical research community. Breast cancer is a common disease among women throughout the world. The medical images and especially Breast Ultrasound (BUS) images are of poor quality, low contrast, and ambiguous. To avoid misdiagnosis, a Computer-Aided Diagnosis (CAD) system has been created for the diagnosis of breast cancer. This study discusses a variety of ultrasonic image segmentation approaches, with an emphasis on several methods developed in the recent four years. As a result, breast ultrasound image segmentation remains a difficult and demanding problem because of several ultrasound aberrations, including strong speckle noise, preprocessing, classification, feature extraction, and segmentation technique to find the accuracy. Lastly, this study outlines the current trends and issues in breast ultrasound images diagnosis, segmentation, and classifications. This review may be useful for both clinicians and researchers who utilize CAD systems for early breast cancer detection.