GC Biopharma has been selected for Korea’s Ministry of Health and Welfare’s Advanced Bio Convergence Talent Development Project to create the world’s first AI-based clinical decision support system designed specifically to forecast hemophilic arthropathy. Currently, standardized models capable of quantitatively assessing the long-term risk of joint deterioration remain limited in practice. This initiative addresses that clinical gap by applying a Hemophilia Joint Damage AI directly into standard treatment workflows.
The development will leverage three decades of accumulated real-world data from Korean hemophilia patients, along with approximately 3,000 specific X-ray images. By utilizing machine learning algorithms, GC Biopharma is constructing a predictive model that analyzes multiple clinical variables. These variables include the patient’s age, historical prophylactic treatment records, and existing levels of joint damage to calculate the future progression of hemophilic arthropathy.
Through this predictive model, healthcare providers will receive actionable data generated by the clinical decision support system, facilitating more precise intervention strategies based on robust AI analytics.
The company has outlined a multi-phase timeline for the Hemophilia Joint Damage AI system:
- Completion of the core predictive model by the end of 2026.
- Development of the X-ray interpretation technology and the initial clinical decision support system prototype next year.
- Finalization of the complete system by 2028, coinciding with patent applications and regulatory approval submissions to the Ministry of Food and Drug Safety.
“Using AI technology, we expect to predict joint damage in hemophilia patients at an earlier stage and support personalized treatment decisions,” stated Choi Bong-gyu, director of GC Biopharma’s AI & Data Science division. The deployment of this technology will serve to manage hemophilic arthropathy more effectively, integrating advanced predictive tools directly into patient care protocols.


















