Breast cancer prediction using a hybrid method based on Butterfly Optimization Algorithm and Ant Lion Optimizer. (December 2021)

• Record Type: Journal Article
• Authors: Thawkar, Shankar; Sharma, Satish; Khanna, Munish; Singh, Law kumar
• Is Part Of: Computers in biology and medicine. Volume 139(2021)
• Publication Date: 2021-12
• Subjects: Butterfly optimization algorithm -- Ant lion optimizer -- Artificial neural network -- Adaptive neuro-fuzzy inference system -- Support vector machine -- Feature selection -- Breast cancer -- Mammography; Medicine -- Data processing -- Periodicals; Biology -- Data processing -- Periodicals; 610.285
• DOI: 10.1016/j.compbiomed.2021.104968
• Languages: English
• ISSNs: 0010-4825

Abstract

Abstract: The design and development of a computer-based system for breast cancer detection are largely reliant on feature selection techniques. These techniques are used to reduce the dimensionality of the feature space by removing irrelevant or redundant features from the original set. This article presents a hybrid feature selection method that is based on the Butterfly optimization algorithm (BOA) and the Ant Lion optimizer (ALO) to form a hybrid BOAALO method. The optimal subset of features chosen by BOAALO is utilized to predict the benign or malignant status of breast tissue using three classifiers: artificial neural network, adaptive neuro-fuzzy inference system, and support vector machine. The goodness of the proposed method is tested using 651 mammogram images. The results show that BOAALO outperforms the original BOA and ALO in terms of accuracy, sensitivity, specificity, kappa value, type-I, and type-II error as well as the receiver operating characteristics curve. Additionally, the suggested method's robustness is assessed and compared to five well-known methods using a benchmark dataset. The experimental findings demonstrate that BOAALO achieves a high degree of accuracy with a minimum number of features. These results support the suggested method's applicability for breast cancer diagnosis. Highlights: Developing a hybrid BOAALO approach for feature selection and breast cancer prediction. The experiment is conducted on 651 mammograms acquired from the digital database for screening mammography (DDSM). The suggested approach is used to predict breast cancer using three classifiers: ANN, ANFIS, and SVM. The methodology is verified against five state-of-the-art approaches utilizing four benchmark datasets. The performance of the proposed approach is evaluated using seven statistical measures.