Efficient Feature Selection Method for Histopathological Images Using Modified Golden Eagle Optimization Algorithm
Surbhi Vijh, Sumit Kumar, Mukesh Saraswat · 2021
Automated medical analysis is growing rapidly for advanced clinical treatment and intervention in diagnosis of patients. Exploration and visualization in the field of histopathology examination are enhancing the decision-making process of pathologists. In this paper, a modified golden eagle optimization technique is proposed for selecting the best subset of features from histopathological images which improve the classification of different categories of histology images. The mean and the best optimal values of the proposed modified golden eagle optimization algorithm is tested on benchmark functions (CEC 2017) namely GEO, WOA, APSO, CSO and LOA. Moreover, the comparative analysis is performed against the recent state of the art algorithm. Furthermore, the experiment results validate that the proposed method outperforms the other considered methods. The accuracy, specificity, sensitivity, recall and F1 score are evaluated for each category of histopathological images using a support vector machine and artificial neural network.