City Research Online

Abstract

Monkeypox, a novel zoonotic disease akin to smallpox, necessitates prompt and accurate diagnosis for effective treatment. Conventional diagnostic techniques, such as Polymerase Chain Reaction (PCR), offer high precision but require specialized equipment and trained personnel, rendering them impractical in resource-limited settings. Existing deep learning approaches for monkeypox diagnosis have predominantly relied on single, resource-intensive models, prioritizing accuracy over deployment feasibility across diverse computing platforms. In this study, we present a resource-conscious model deployment strategy that balances diagnostic accuracy with computational efficiency, enabling precise diagnosis in both resource-rich environments, such as hospitals, and resource-constrained contexts. We utilize a pretrained ConvNeXt-B model, trained on the extensive ImageNet-22K dataset, for deployment in resource-abundant scenarios, and a SqueezeNet model, optimized for resourcelimited devices using the ImageNet-1K dataset. To enhance the performance of the lightweight SqueezeNet model without increasing computational complexity, we apply Knowledge Distillation. Experimental results demonstrate that the ConvNeXt-B model achieves an accuracy of 95.75%, which is 3.47% higher than the previous studies. Similarly, the Knowledge Distillation-enhanced SqueezeNet model attains an accuracy of 91.89%, representing a 2.32% improvement over the baseline. This dualmodel approach ensures that accurate monkeypox diagnostics are accessible across a wide range of computational environments, thereby supporting more effective outbreak management and contributing to improved public health outcomes.

Publication Type:	Conference or Workshop Item (Paper)
Additional Information:	Copyright: 2025, IEEE.
Publisher Keywords:	Accuracy, Hospitals, Computational modeling, Telemedicine, Rendering (computer graphics), Skin, Security, Polymers, Public healthcare, Diseases
Subjects:	Q Science > QA Mathematics > QA75 Electronic computers. Computer science R Medicine > RA Public aspects of medicine > RA0421 Public health. Hygiene. Preventive Medicine
Departments:	School of Science & Technology School of Science & Technology > Computer Science
SWORD Depositor:	Symplectic Administrator

Export

Downloads