Utilizing convolutional neural networks to classify monkeypox skin lesions.

Area of Science:

• Medical Informatics
• Artificial Intelligence in Medicine
• Computational Biology

Background:

• Monkeypox is a rare viral disease causing severe illness with characteristic skin lesions.
• Accurate visual diagnosis of monkeypox lesions is challenging, especially in resource-limited settings.
• Deep learning, specifically Convolutional Neural Networks (CNNs), shows promise for image classification tasks.

Purpose of the Study:

• To develop and evaluate a deep learning model for classifying monkeypox skin lesions.
• To optimize the CNN model's performance using the Grey Wolf Optimizer (GWO) algorithm.
• To assess the potential of the optimized model for improving monkeypox diagnosis and public health surveillance.

Main Methods:

• Implementation of Convolutional Neural Networks (CNNs) for image classification of monkeypox skin lesions.
• Optimization of the CNN model architecture and parameters using the Grey Wolf Optimizer (GWO) algorithm.
• Performance evaluation using metrics including accuracy, precision, recall, F1-score, and AUC.

Main Results:

• The optimized CNN model achieved a high accuracy of 95.3% in classifying monkeypox skin lesions.
• The Grey Wolf Optimizer (GWO) significantly improved the model's discriminative ability compared to the non-optimized model.
• Enhanced performance metrics (accuracy, precision, recall, F1-score, AUC) were observed with GWO optimization.

Conclusions:

• The GWO-optimized CNN model offers a highly accurate and efficient approach for monkeypox diagnosis.
• This AI-driven method has the potential to facilitate earlier detection and improve patient outcomes.
• The approach holds significant public health implications for controlling and preventing monkeypox outbreaks through enhanced surveillance.