Modelling Monkeypox Outbreak Using Machine Learning

Abstract

Monkeypox is an emerging zoonotic disease that has gained global attention due to its rapid spread beyond traditionally endemic regions. Early diagnosis plays a critical role in controlling outbreaks and ensuring timely treatment. However, Monkeypox lesions often resemble those of Chickenpox, Smallpox, and other viral infections, which complicates manual diagnosis. This paper proposes a hybrid machine learning framework that combines deep learning-based image classification with text-based symptom analysis. DenseNet121 was applied to skin lesion classification, while TF-IDF with Logistic Regression was used for text features. Both models were integrated into a Flask-powered backend and a lightweight web interface for real-time predictions. Our study reports an accuracy of 90.91% for DenseNet121 on Images, 98.03% for the text classifier, and 94.19% for both. These results demonstrate the promise of multimodal AI systems in supporting clinicians during outbreaks.