Optimizing Skin Disease Classification: A Comprehensive Investigation of CNN-SVM Hybrid Models with Layer Configurations

This research examines skin disease classification in depth using a hybrid approach that combines Convolutional Neural Networks (CNN) and Support Vector Machines (SVM). The model architecture is made up of three convolutional layers, three max-pooling layers, and two fully connected layers, all of which have been painstakingly explored for optimal configurations. The study focuses on common skin disorders such as eczema, psoriasis, acne vulgaris, and rosacea, with a focus on precision, recall, and F1-score metrics. The layer examination dives into crucial parameters including filter widths, kernel types, and activation functions, exposing how they affect model performance. The results show that the proposed CNN-SVM model is robust, with a weighted average accuracy of around 94.12%. SVM integration improves classification, highlighting the synergy between deep learning and traditional artificial intelligence approaches. The present research not only advances the precision of skin condition classification, but it also gives useful information for healthcare practitioners and researchers. The findings highlight the importance of sophisticated layer configurations as artificial intelligence continues to transform medical diagnoses. Future research areas include optimization modifications, new datasets, and real-world deployment concerns, all of which will help to confirm the model's efficacy in improving dermatological diagnosis. This research contributes to the increasing convergence of technology and healthcare, paving the way for the use of machine learning for improved skin disease diagnostics.