Select Language:
Chinese tech giants like Alibaba and Tencent are making significant progress in integrating artificial intelligence into critical medical care. After multiple upgrades, these AI models are now moving beyond support roles and actively participating in diagnosing and treating seriously ill and emergency patients.
Recently, Alibaba’s research arm, in collaboration with the First Affiliated Hospital of Zhejiang University School of Medicine, introduced a large language model called iAorta. This tool is designed to assist in chest pain emergencies. Using a standard, non-contrast CT scan, physicians can now detect acute aortic syndromes within seconds, reducing the diagnostic process to less than two hours. This breakthrough was detailed in a recent publication in a prominent medical journal.
Meanwhile, Tencent has developed its own critical care AI model, known as the Qiyuan Critical Care Model, in partnership with Shenzhen Mindray Bio-Medical Electronics. It is the world’s first AI system tailored for intensive care units and has begun deployment across several hospitals. This technology assembles a patient’s comprehensive medical profile in just five seconds, forecasts potential health trends through advanced reasoning, and offers treatment recommendations.
Previously, AI tools in medicine were mostly confined to virtual assistance or increasing the efficiency of medical staff—such as automating the generation of medical records. As of May, nearly 300 medical-focused large language models (LLMs) had been released in China, with over half actively supporting healthcare services.
Both the iAorta for emergency chest pain diagnosis and the Qiyuan model for intensive care are now being utilized in critical medical scenarios—the core functions of diagnosis and patient treatment.
Zhang Hongkun, director of the Vascular Surgery Department at Zhejiang University’s First Hospital, emphasized the importance of these advancements: “Most patients with acute aortic syndrome experience nonspecific symptoms like mild chest or abdominal pain, leading to a diagnosis miss rate of around 40%. With AI’s help, doctors can quickly run a simple CT scan to identify aortic dissection, which is crucial for effective treatment.” He noted that approximately 40 to 50 percent of these patients die within 48 hours if not diagnosed promptly, underscoring the need for rapid, precise detection.
At the 2025 World Artificial Intelligence Conference this July, LAN Holding, a leading AI healthcare provider, introduced its flagship product—the Digital Twin Clinical Platform and LAN-AI Agent for Critical Care. This system operates in high-stakes hospital environments, including emergency rooms, surgical theaters, and intensive care units.
By inputting vital signs, lab results, imaging data, medical records, and device information, the LAN-AI Agent generates a “digital twin” of the patient. Healthcare providers can continuously monitor this virtual representation, assess potential risks, and receive AI-driven diagnosis and treatment plans.
These cutting-edge AI models are currently being tested in some of China’s top hospitals, such as Fudan University’s Huashan Hospital and the Shanghai Municipal Hospital of Traditional Chinese Medicine.
Add us on ChatGPT
Add us on Perplexity
