GuShiio.com ① Following the release of AI pediatricians at Beijing Children’s Hospital, another large pathology model has been launched in Shanghai, which means that the development of AI diagnosis and treatment in my country is accelerating.
② Data shows that the size of China’s AI medical market has increased from 2.7 billion yuan in 2019 to 8.8 billion yuan in 2023, and is expected to be close to 30 billion yuan in 2028.
Cailian News, February 18 (Reporter Zheng Bingxun) Following the official application of AI pediatricians at Beijing Children’s Hospital in clinical practice five days ago, another large AI model in the medical field was released today, and the development of AI diagnosis and treatment in my country is accelerating.
Today, at the “2025 Medical Artificial Intelligence and Precision Diagnosis and Treatment Development Forum” held by Ruijin Hospital affiliated to Shanghai Jiao Tong University School of Medicine, Ruijin Hospital and Huawei jointly released a large pathology model-RuiPath Path.
It is understood that in cancer diagnosis and treatment, early detection, early diagnosis and early treatment are crucial. However, due to shortcomings such as the huge gap in pathologists and unbalanced distribution, the compliance rate for first diagnosis in primary hospitals in my country is low. The emergence of large pathological models has become an important force in solving this dilemma.
At today’s forum, Ning Guang, academician of China Academy of Engineering and president of Ruijin Hospital, introduced that RuiPath, a large pathology model, was developed based on the full business process of the Digital Intelligent Pathology Department of Ruijin HospitalClinic-level domestic multi-modal interactive pathology modelDigitalization and AI help hospitals achieve multimodal integration.
Wang Chaifu, director of the pathology department of Ruijin Hospital, gave an example to illustrate the difference between RuiPath and traditional diagnosis and treatment methods.”In the interactive auxiliary diagnosis process, the traditional diagnosis method is for the doctor to view the slices one by one under the microscope, and then manually enter the report after completing the diagnosis.RuiPath can accurately identify the lesion area in advance, and the single-slice AI diagnosis time only takes a few seconds.”
Wang Chaofu said that this changed the doctor’s working mode from looking for lesions one by one under the microscope to reviewing AI diagnosis results in an interactive manner, and upgraded from the “Slide by slide” mode to the “Step by step” mode., significantly improving the efficiency and quality of diagnosis.
In addition, in terms of the depth of knowledge of RuiPath, take a sub-specialist pathologist as an example. In the first 10 years of study, he needed to study at least 50 relevant professional books and learn to diagnose 500,000 pathological slices. In just two months of research and development, RuiPath has “studied” more than 300 pathological diagnosis books and “read” 1 million digital slices.
Da Qian, a pathologist at Ruijin Hospital, further introduced,”In terms of breadth,RuiPath covers 90% of cancer types suffered by people with cancer in China every year。In terms of depth, the depth of questions and answers on sub-specialty knowledge reaches the expert level. In the test of commonly used questions compiled by pathologists,RuiPath’s answer accuracy rate is over 90%, and is at the leading level at home and abroad in graphic and text question and answer tasks in medical examination scenarios.”
Before the release of RuiPath, the “AI Pediatrician” jointly developed by Beijing Children’s Hospital and Baichuan Intelligent was launched on January 18 and officially put into clinical application on February 14. It is understood that in a previous consultation around an 8-year-old child, the AI pediatrician and 13 well-known experts from different departments such as otolaryngology head and neck surgery and oncology surgery gave highly consistent suggestions.
In the eyes of senior medical staff, AI pediatricians can not only serve as clinical research assistants, helping doctors quickly obtain the latest scientific research results and authoritative guidelines, but also assist doctors in the diagnosis and treatment of difficult and rare diseases and improve clinical decision-making efficiency.
Ning Guang believes thatAI technology will reshape the operating efficiency and accuracy of hospitals, and the future of the entire society will also be reshaped by AI and big models.If you don’t hug you will become an outsider. “In the near future, the integration of medical scenarios and digital equipment will produce more new medical technologies with humanistic perspectives and sentiments, and realize broader application prospects of multimodal fusion technology in the medical field.” he said.
In fact, in recent years, China has successively introduced multiple policies to promote the development of AI medical care.
Including,In November 2024, the National Medical Insurance Administration will include AI-assisted diagnosis technology in the establishment guidelines for medical service price projectsIn the same month, the National Health Commission and other departments jointly issued the “Reference Guidelines for Artificial Intelligence Application Scenarios in the Health Industry”, which clarified the basic concepts and application scenarios in more than 80 sub-areas such as intelligent assisted diagnosis in medical imaging. The following month, the 2024 China Medical Artificial Intelligence Conference was held in Shanghai, launching the China Artificial Intelligence Medical Device Innovation Industry Alliance.
The scale of China’s AI medical market is also continuing to expand. Huaxi Securities quoted third-party data as showing that from 2019 to 2023, the size of China’s AI medical market grew rapidly from 2.7 billion yuan to 8.8 billion yuan, and the market size is expected to be close to 30 billion yuan in 2028.
It is worth mentioning that since the beginning of this year, DeepSeek has set off a wave of change in various fields, and the medical field is no exception.According to incomplete statistics, in addition to Hengrui Pharmaceutical’s announcement of access to DeepSeek, leading companies in the digital medical field such as Ark Jianke, Yidu Technology, Yingtong Airdoc, and Wanda Information have successively announced access to DeepSeek.
Soochow Securities bluntly stated that AI has empowered many aspects of medical services. Among them, in terms of improving medical efficiency and quality, AI can help the rapid accumulation of medical knowledge bases and promote continuous learning and iteration of models, reduce doctors ‘repetitive workload and allow doctors to focus on clinical practice., realize standardization of medical treatment processes to reduce human operation risks, and achieve remote consultation in primary medical institutions to improve service coverage.
In addition, areas where AI medical applications are relatively mature include imaging (pathological imaging, endoscopic imaging and other types of imaging), as well as fields such as pharmaceuticals, robots, clinical decision-making systems, genetic testing, and intelligent pathology.
