On April 28, at the RIMAG GROUP (02522) strategic upgrade and global digital intelligence imaging⁺ ecosystem conference, the company officially launched a milestone product: the Brain CT Super Intelligent Agent (Dr. Xiao Jun 2.0). This system was jointly developed by RIMAG GROUP's incubated AI enterprise, Yinghe Yimai, and Beijing Tiantan Hospital affiliated with Capital Medical University. It is recognized as the world's first medical imaging AI-assisted diagnostic system to achieve complete, project-level intelligence. Liu Yaou, Director of the Medical Imaging Department at Beijing Tiantan Hospital, conducted a live demonstration of the product's real-world application within the hospital. Trained on nearly 300,000 cases, over ten million image-text pairs, and corresponding labels, the Brain CT AI covers 94 common brain diseases and has achieved a comprehensive diagnostic accuracy rate of 87.8%. In tests at Tiantan Hospital, over 90% of cases required no manual corrections. The system generates structured reports directly from images, compressing the review process from 15 minutes to just one minute, truly realizing the goal of "One System, One Model, One Minute, One Report." Meng Tao, the head of the artificial intelligence business, stated that the product is supported by the three foundational pillars of Yinghe Yimai's "Hetu Plan": standard infrastructure, data infrastructure, and model infrastructure, along with an AI manufacturing model based on "working is training, using is evolving." He revealed that, guided by a philosophy of local iteration, the Brain CT Super Intelligent Agent has undergone several months of closed-loop refinement within Tiantan Hospital. The next step involves rapidly replicating the technology for other areas such as abdominal CT, knee joint MR, and aortic CTA, with plans to complete 12 industry standards and corresponding datasets by the end of the year. This launch signifies that the strategic collaboration between RIMAG GROUP and Yinghe Yimai has progressed from technological incubation to product commercialization, setting a new industry benchmark for medical imaging AI's evolution from a "single-point tool" to "full-disease, full-process intelligence."
Comments