<![CDATA[The rising number of skin cancer cases in recent decades and the limitations of early detection methods, such as costly and less accessible biopsies, highlight the need for an affordable, accessible solution. This study aims to design a smart mobile healthcare application integrating a Convolutional Neural Network (CNN) to detect skin cancer early through digital imaging. Using a dataset of 5,100 images categorized into melanoma, non-melanoma, and normal skin, the CNN model based on VGG16 architecture was trained and evaluated using accuracy, precision, recall, and F1-score. The model achieved 93.14% testing accuracy, 86.93% average training accuracy, and the F1-score 0.89. The UI/UX design follows the design thinking approach, emphasizing a user-friendly, fast, and interactive interface. Core features include user login, skin image scanning, classification results, and AI-based consultation. The application is intended to serve as an effective, accessible tool for early skin cancer detection, supporting timely clinical diagnosis for all users.]]>
Copyrights © 2025
