An Intelligent Artificial Bee Colony and Adaptive Bacterial Foraging Optimization Scheme for Reliable Breast Cancer Diagnosis

Abstract

: Breast cancer is essential to be detected in primitive localized stage for enhancing the possibility of survival since it is considered as the major malediction to the women society around the globe. Most of the intelligent approaches devised for breast cancer necessitate expertise that results in reliable identification of patterns that conclude the presence of oncology cells and determine the possible treatment to breast cancer patients in order to enhance their survival feasibility. Moreover, the majority of the existing schemes of the literature incur intensive labor and time, inducing a predominant impact over the diagnosis time utilized for detecting breast cancer cells. An Intelligent Artificial Bee Colony and Adaptive Bacterial Foraging Optimization (IABC-ABFO) scheme is proposed for facilitating a better rate of local and global searching ability in selecting the optimal features subsets and optimal parameters of ANN considered for breast cancer diagnosis. In the proposed IABC-ABFO approach, the traditional ABC algorithm used for cancer detection is improved by integrating an adaptive bacterial foraging process in the onlooker bee and the employee bee phase that results in optimal exploitation and exploration. The investigation of results of the proposed IABC-ABFO approach facilitating the use of the Wisconsin breast cancer dataset showed a mean classification accuracy of 99.52% which is higher than the existing breast cancer detection schemes.