Classification of abnormalities in breast ultrasound images using ANN, FIS and ANFIS classifier: A comparison

Abstract

Abstract When diagnosed early, cancer can be cured. The leading cause of death for women is breast cancer, the most common form of cancer in women. However, there are several ways to think about breast cancer, the US is widely used as it is not aggressive and painless. The appearance of dots on ultrasound images reduces image clarity, affecting image quality. Dots are drawn to accurately define distorted images. This is accompanied by the separation of the wound using a level setting method, after which the features of the lesions are removed. Only the most distinct traits are considered. Using the division of ANN, FIS and ANFIS, with a accuracy of 92.3 percent, 88 percent and 96 percent, the inconsistencies in the 41 images of Breast Ultrasound are identified and classified.