MonkeyPox skin disease classification

Abstract

To stop the spread of monkeypox, a viral skin disease and other skin diseases, early detection and precise classification are essential. Using VGG16, ResNet50, CNN, and AlexNet, this work investigates a Machine Learning (ML)-based system for categorizing monkeypox and other skin conditions. We trained many deep learning models, did data augmentation, and downloaded lesion images from Kaggle. custom CNN (94.66%) and VGG16 (80.26%) attained the best accuracy , whereas ResNet50 (55.33%) and AlexNet (37.35%) had poorer stability. We created a React JS web application that allows users to contribute photographs for analysis in real-time classification. In order to increase model performance and interpretability, future developments will incorporate Explainable AI (XAI), GPU acceleration, and dataset expansion. Our study shows how ML-driven skin disease identification can be used for early diagnosis and public health monitoring.