Answers are grounded in this paper and cite their sources.
Background: The integration of computer science and medicine has led to the development and broad application of intelligent disease detection models based on deep learning and convolutional neural networks. In this study, we aimed to test the capacity of an enhanced lightweight YOLOv8n-cls model to improve the recognition of skin lesions caused by mpox (previously called monkeypox) and other diseases. The goal was to develop a model with greater accuracy and efficiency than existing baseline models. Methods: The proposed model incorporates an enhanced multi-scale convolutional pathway and the Convolutional Block Attention Module into the backbone network. The model’s performance was evaluated using four metrics (accuracy, precision, recall, and F1 score) and compared to that of ResNet50, YOLOv11n-cls, EfficientNet_b0, and DenseNet169. Grad-CAM++ heatmaps were generated for visual analysis. Results: The model achieved 97.9% accuracy, 98.2% precision, 98.0% recall, and an F1 score of 0.98. It outperformed ResNet50 (F1 score: +3.8%) and YOLOv11n-cls (F1 score: +7.100%). The training time was reduced by 60%-72% compared to that recorded for DenseNet169. The results of our ablation experiments validated the contributions of the individual modules. Conclusion: This lightweight YOLOv8n-cls model, which was augmented with a multi-scale pathway and an attention mechanism, demonstrated significantly improved recognition performance at a reduced computational cost. Our findings advance the development of efficient, accurate, artificial intelligence-driven disease detection systems for use in healthcare settings.
No takes yet. Share an insight, caveat, or question.
A 2025 study studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: