Optimization of Feature Selection Using Greylag Goose Optimization Algorithm for Monkeypox

Monkeypox is an illness like smallpox that began to spread through several countries at a relatively rapid pace. The rash is among monkeypox's most outstanding clinical features; however, a similar rash is evident in measles and chickenpox patients as well. AI and computer vision are well on their way to becoming must-have medical tools. For instance, computer-aided design (CAD) uses visual data to diagnose diseases such as chickenpox and measles at their early stage. Proposing a similar utilization of the AlexNet pre-trained model in extracting the differential features from MSID, the research has recorded an impressive precision rate of 0.932295, a testament to the credibility and precision of our research. We apply feature selection to reduce the extracted features in our proposed binary Greylag Goose Optimization (bGGO) method, a novel approach that we believe has the potential to significantly outperform existing models. It gives a better average fitness of 0.60068 and fixed best fitness as 0.50248. The presented model, with its novel approach, is discussed with several other optimization models, namely, binary waterwheel plant algorithm (bWWPA), Boosted Dipper Throated Optimization (bDTO), binary particle swarm optimizer (bPSO), binary whale optimization algorithm (bWAO), binary gray wolf optimizer (bGWO), and binary firefly algorithm (bFA). For the possibility of a difference between the subjects in the suggested approach and other methods, the results were subjected to the Wilcoxon signed-rank test and Analysis of variance. This comparison supported the novelty of this proposed method for the identification of monkeypox, sparking interest in its unique approach.