About
About Dermasight
An undergraduate thesis project applying
computer vision to skin lesion analysis
What is Dermasight
Dermasight is a web-based application that uses computer vision and deep learning to analyze images of skin lesions. Developed as an undergraduate thesis project, it aims to explore how artificial intelligence can support early skin health awareness by providing users with an accessible, instant image-based assessment tool.
The system is trained on publicly available dermoscopic image datasets and applies convolutional neural network techniques to classify and evaluate skin lesion characteristics. Beyond classification, Dermasight integrates large language models (LLMs) grounded in expert dermatological guidelines and clinical literature to generate clear, objective, and relevant descriptive information.
Dermasight is intended to serve as a proof-of-concept for accessible, AI-assisted skin screening — not as a replacement for professional medical diagnosis.
Project Background
This project was developed as part of the undergraduate thesis requirement at the University of St. La Salle - Bacolod, College of Computing Studies. It reflects the researchers' exploration of real-world applications of computer vision, particularly in the context of healthcare accessibility and early detection.