Abstract

Skin diseases represent a significant global health concern, requiring early and accurate diagnosis to prevent complications and make effective treatment. Manual clinical examination is often time-consuming and dependent on expert dermatological knowledge, which may not always be accessible in remote or resource-limited regions. This paper proposes an automated skin disease detection and classification system using the Inception-V3 deep learning architecture integrated with a Flask-based web application. The Inception-V3 model pretrained on large scale image datasets and fine-tuned using transfer learning, enables efficient multi-class classification of different skin conditions from dermoscopic images. Image preprocessing techniques such as resizing, normalisation, and data augmentation are applied to enhance model robustness and reduce overfitting. The trained model achieves high classification accuracy, precision, recall, and F1-score, demonstrating its effectiveness in distinguishing between different skin disease categories. To ensure practical usability, the system is deployed through a user-friendly Flask web interface that allows users to upload skin images and receive real-time diagnostic predictions. The proposed framework provides a scalable, cost-effective, and accessible solution for preliminary skin disease screening, supporting telemedicine and assisting healthcare professionals in early-stage diagnosis and decision-making processes.