Computer Aided Breast Cancer Diagnosis using Arithmetic Optimization Algorithm with Fuzzy Logic on Histopathological Images

Computer-aided breast cancer (BC) detection is a medical application that utilizes machine learning approaches and computer algorithms to support doctors in the diagnosis and detection of BC. The primary target is to enhance the efficiency and accuracy of BC diagnosis and screening. BC detection on histopathological images (HPI) is the process of analyzing tissue samples collected during surgery or biopsy to identify cancer cells. By examining these tissue samples under a microscope HPI is acquired and used for the BC grading and diagnosis. This study presents a new Computer Aided Breast Cancer Diagnosis by employing an Arithmetic Optimization Algorithm with Machine Learning (CABCD-AOAML) approach on Histopathological Images. The presented CABCD-AOAML approach intends to detect and classify the occurrence of BC. In the initial phase, the CABCD-AOAML technique applies the CLAHE model for the contrast enhancement process. Besides, the CABCD-AOAML technique uses an ensemble feature fusion process, comprising residual network (ResNet) and DenseNet model. To adjust the hyperparameters of the DL models, the AOA is used to improve the performance. Lastly, the adaptive neuro-fuzzy inference system (ANFIS) approach is employed for the recognition and classification of BC. The investigational output of the CABCD-AOAML technique is assessed on the BreakHis database. The attained outputs portray the better achievement of the CABCD-AOAML technique over recent methods.