Challenge Grant: Image-based Detection for Rheumatoid Arthritis with Synthetic Data

This research aims to develop a machine learning model that estimates the presence of inflammation caused by rheumatoid arthritis from hand images captured with a standard camera, with the goal of enabling early detection and supporting telemedicine applications. To address the scarcity and imbalance of medical image data, we generate synthetic data that reproduces the hands of rheumatoid arthritis patients and use these data to improve estimation accuracy. By establishing a diagnostic technology that does not require specialist physicians or expensive equipment, the proposed approach is also expected to contribute to reducing disparities in access to healthcare.