SELO: Smack Echolocation Optimization A Novel Approach for Breast Cancer Thermogram Classification

This research presents the advanced method of classification of breast cancer thermograms using a self-attenuated deep Convolutional Neural Network (CNN). Meta-heuristic and evolutionary computation is popular in nature-inspired optimization methods. The classification model is trained using the novel Smack Echolocation Optimization (SELO) algorithm, which combines Dolphin Echolocation and Jellyfish searching features to improve the classification efficiency. The SELO algorithm shows better convergence speed and parameter tunability, which leads to improved classification accuracy. The SELO algorithms phases are based on the jellyfish search movements and dolphin echolocation behavior which ensure the discovery of the best solution. Comparative analysis of the existing methods shows the effectiveness of optimized SELO self-attenuated deep CNN. The precision, recall, and F1 measures for breast cancer detection are significantly enhanced. This study provides compelling evidence for the suitability of bio-mimicking optimization methods in medical imaging and shows the significance of early diagnosis of breast cancer.