Pre-trained Deep learning model for Monkeypox Prediction using Dermoscopy Images in Healthcare

Abstract

Monkeypox is a medical skin problem that can be transferred from animals to humans and then from one person to other. Its species is Otho poxvirus. The manifestations of monkeypox and smallpox are virtually identical thus, antiviral medication developed to prevent the smallpox virus may be used for monkeypox despite the absence of effective therapy. Infected individuals, smallpox vaccination, prevention infection, and use of personal Protective Equipment (PPE) kits are all part of the control of monkey pox. In this study, deep learning-based convolution neural network (CNN) is used to detect monkeypoxes. In this research, three optimizers namely SGD, RMSProp and Adam are employed to predict monkeypox. From the three optimizers, the best optimizer is selected based on accuracy. The SGD optimizer achieves highest accuracy of 93.39% in 100 epochs. Other optimizers were RMSProp and Adam, with scores of 91.30% and 93.22%, respectively. Using a single image of an infected person, the CNN model easily predicts the monkeypox virus. This model can be used as second source of opinion for medical practitioners to identify the monkeypox.