Computer-Aided Detection and Classification of Monkeypox and Chickenpox Lesion in Human Subjects Using Deep Learning...

Area of Science:

• Medical Imaging
• Artificial Intelligence in Dermatology
• Infectious Disease Diagnostics

Background:

• Monkeypox outbreaks necessitate rapid and accurate diagnostic methods.
• Distinguishing monkeypox from chickenpox is challenging due to similar skin lesions.
• Misdiagnosis can lead to disease spread and outbreaks.

Purpose of the Study:

• To develop and evaluate a deep learning model for early detection and classification of monkeypox skin lesions.
• To differentiate monkeypox from chickenpox using digital images.
• To improve diagnostic accuracy and reduce the risk of future outbreaks.

Main Methods:

• Utilized open-sourced digital images of monkeypox and chickenpox skin lesions.
• Developed a two-dimensional convolutional neural network (CNN) with four convolutional and three MaxPooling layers.
• Compared the proposed CNN model against state-of-the-art deep learning models.

Main Results:

• The proposed CNN model achieved a test accuracy of 99.60%.
• Achieved a weighted average precision, recall, and F1 score of 99.00%.
• Outperformed other deep learning models, including AlexNet (98.00%) and VGGNet (80.00%).

Conclusions:

• The developed CNN model demonstrates high accuracy and generalization for detecting monkeypox skin lesions.
• The model effectively avoids overfitting through unique architecture and image augmentation.
• This deep learning approach offers a valuable tool for rapid and accurate monkeypox diagnosis from digital images.