Enhanced Skin Disease Detection and Classification using Deep Learning
Skin ailments are a global health concern, impacting millions and requiring timely diagnostic precision to prevent complications. This project, titled "Enhanced Skin Disease Detection and Classification using Deep Learning," introduces an automated framework for the preliminary screening of dermatological conditions. Developed using a robust Python-Flask architecture, the platform features a web-based interface that allows users to upload clinical images for rapid evaluation.
The technical core of the system utilizes two state-of-the-art Convolutional Neural Networks (CNNs): MobileNet and InceptionV3. These models were trained on a specialized dataset to recognize and categorize nine distinct labels, including bacterial infections like Cellulitis and Impetigo, fungal conditions such as Ringworm and Nail Fungus, and viral outbreaks like Chickenpox and Shingles.
Experimental results highlight the exceptional performance of the MobileNet model, which achieved a 97% test accuracy, outperforming InceptionV3 in terms of precision and recall. Beyond mere classification, the system offers a complete diagnostic report, including visual data distribution charts and suggested care tips. This project stands as a significant advancement in medical AI, providing an accessible tool for early skin disease assessment and clinical decision support.