A Comparative Analysis of CNN-Based Pretrained Models for the Detection and Prediction of Monkeypox

Following the COVID-19 epidemic, Monkeypox, which is an uncommon illness, elicited concerns from specialists in the medical field. It is a cause for concern because Monkeypox is difficult to diagnose in its early stages, as its symptoms may resemble those of measles and chickenpox. In addition, there is a knowledge gap among healthcare experts regarding this rare ailment. As a consequence, there is a pressing need to develop a new method to combat and predict the disease at the initial stages of viral infection. This research included a variety of pretrained convolutional neural network (CNN) models. The models that were used included VGG-16, VGG-19, Resnet50, Inception-V3, DenseNet, Xception, MobileNetV2, Alexnet, LeNet, and majority Voting. To conduct this study, several different datasets were integrated, including the following: Monkeypox against chickenpox, Monkeypox versus measles, Monkeypox versus normal, and Monkeypox versus all disorders. In the case of Monkeypox versus chickenpox, majority voting scored 97%, while Xception scored 79% for Monkeypox versus measles, MobileNetV2 scored 96% for Monkeypox versus normal, and LeNet scored 80% for Monkeypox versus all.