AI-Driven Skin Cancer Detection for Early Intervention using Deep Learning

Skin cancer is a significant global health issue, with increasing incidence rates posing a challenge to healthcare systems. This research presents an AI-driven system for skin cancer detection and patient prioritization, aimed at addressing the shortage of dermatologists and facilitating early intervention. Utilizing the HAM10000 dataset, we trained an Xception convolutional neural network to classify skin lesions into seven distinct categories: Actinic Keratosis (akiec), Basal Cell Carcinoma (bcc), Benign Keratosis (bkl), Melanoma (mel), Dermatofibroma (df), Melanocytic Nevus (nv), and Vascular Lesion (vasc). The system features a user-friendly interface for image upload, automated segmentation, and classification prediction, accompanied by a downloadable PDF report. We emphasize the system's role as a prioritization tool, not a definitive diagnostic solution. The model achieved an accuracy of 89% on the test dataset, demonstrating its potential to enhance early detection and streamline patient management in dermatology.