On the morning of February 19, a pioneering research team, spearheaded by Professor Sun Kun and Professor Yu Yongguo from Xinhua Hospital, affiliated with the Shanghai Jiao Tong University School of Medicine, joined forces with Professor Zhang Ya and Associate Professor Xie Weidi from Shanghai Jiao Tong University, to introduce DeepRare—the world's inaugural AI-powered diagnostic system for rare diseases, distinguished by its 'traceable reasoning process.' The groundbreaking accomplishments were promptly published in the esteemed international journal Nature on the same day.

This innovative system adeptly mimics the diagnostic reasoning of human physicians, employing a methodical approach of hypothesis formulation, verification, and self-correction. Each diagnosis is meticulously supported by a comprehensive 'chain of evidence.' Testing has revealed that, relying solely on clinical symptoms, DeepRare achieves an impressive initial diagnostic accuracy rate of 57.18%, outperforming the leading international models by nearly 24 percentage points. When genetic data is integrated, its accuracy soars to over 70%.

Presently, DeepRare has been integrated into an online diagnostic platform, attracting registrations from over 600 medical institutions globally. Additionally, it is set to be officially implemented at Shanghai Xinhua Hospital, providing invaluable support to doctors in minimizing the risk of missed diagnoses. The research team is also poised to establish a global alliance dedicated to the diagnosis and treatment of rare diseases, with plans to rigorously validate the system using 20,000 real-world cases within the upcoming six months.