ARCHIVES

Skin diseases are among the most prevalent medical conditions worldwide, affecting millions of people across all age groups. Early and accurate diagnosis of skin conditions is essential to prevent serious health consequences, including skin cancer. However, dermatological expertise is limited in many regions, making access to timely diagnosis a significant challenge. This project presents DermaAI, an AI-powered skin disease classification system that leverages deep learning techniques to automatically identify and analyze common skin conditions from uploaded medical images The proposed system uses EfficientNet-B0 with transfer learning as the core deep learning architecture. The model is trained on a comprehensive dataset combining the DermNet NZ Image Library and the ISIC 2019 Challenge Dataset, enabling it to classify 23 distinct skin conditions including acne, melanoma, psoriasis, eczema, basal cell carcinoma, and tinea. The system incorporates a robust preprocessing pipeline consisting of image resizing to 224x224 pixels, normalization using ImageNet mean and standard deviation, and extensive data augmentation techniques including rotation, horizontal flipping, color jitter, and brightness adjustment.The DermaAI system is deployed as a Flask-based web application providing an intuitive drag-and-drop image upload interface with real-time prediction results, confidence scores, and detailed medical information. The system achieves an overall classification accuracy of 95.6% on the test dataset with an average prediction time of less than 2 seconds per image. Additional features include skin detection validation using HSV color space analysis, top-3 prediction display, treatment recommendations, and prominent medical disclaimers encouraging professional consultation.Experimental results demonstrate that the proposed system provides highly accurate, fast, and accessible dermatological screening support. The system has strong potential to improve healthcare accessibility, particularly in regions with limited dermatological services, while contributing to early detection of serious skin conditions