Whale Optimization Algorithm Based Improved Convolutional Neural Network for Breast Cancer Detection and Diagnosis

The breast cancer is a hazardous disease through the huge mortality and morbidity rate and its detection id identified as important health issues in nowadays. Numerous studies have conducted to categorize the patients into benign and malignant but the significance of the issues and efforts are continuing. This paper proposed a Whale Optimization Algorithm based Improved Convolutional Neural Network (WOA-ICNN) for breast cancer detection and diagnosis. The BCWD dataset is utilized and preprocessed through data normalization. Then, the features are extracted by Grey Level Cooccurrence Matrix (GLCM) and selected by WOA. The selected features are given to ICNN for early detection and diagnosis of breast cancer. The accuracy, precision, recall and f1-score are utilized for estimating WOA-ICNN performance. The WOA-ICNN obtains 99.16% accuracy, 98.75% precision, 98.87% recall and 98.66% f1-score which is better than existing algorithms like Fast-Learning Network (FLN), optimized Deep Recurrent Neural Network (RNN), Multi-Layer Perceptron (MLP) based homogeneous ensemble.