FURI | Spring 2025

Utilizing Machine Vision for Accurate Identification and Classification of Skin Diseases

The project explores machine vision applications for accurate and unbiased skin disease identification. Using a CNN model trained with the Skin Diseases Image Dataset, the project aims to develop a model with high classification accuracy for various skin diseases. Data augmentation will improve the generalization of various skin tones and combat diagnostic bias. The model’s performance will be evaluated using accuracy, precision, recall, and fairness metrics. The outcome is an effective diagnostic tool that promotes equitable healthcare by boosting accessibility and accuracy in skin disease detection.