Enhancing Monkeypox Skin Lesion Detection: A Fusion Approach of VGG 16 and Xception Features with SVM Classifier

Monkeypox, a viral ailment resembling smallpox, can be identified through transfer learning, which involves utilizing pre-trained deep learning models to recognize patterns in medical images and facilitate early detection. This study assesses the efficacy of pre-trained convolutional neural network (CNN) models such as VGG 16, VGG 19, InceptionV3, and Xception as feature extractors. The study combines non-handcrafted features from these models, creating a final feature matrix that is inputted into various conventional machine learning classifiers. Testing was conducted on a publicly available dataset of monkeypox skin images, with the best performance achieved by VGG 16 + Xcepetion + SVM, exhibiting an accuracy of 97.14%, a sensitivity of 93.75%, and a specificity of 100%. This research highlights the potential of deep learning in medical image analysis and its potential to aid clinicians in the early detection of monkeypox.