HyperAIHyperAI
Back to Headlines

AI-Powered Medical Training Revolutionizes Education

5 days ago

On September 25, 2025, Professor Yaqin Zhang, a Foreign Member of the Chinese Academy of Engineering, Chair Professor at Tsinghua University, and Director of the Institute for AI Industry Research (AIR) at Tsinghua University, co-authored a landmark article published in the prestigious international journal Nature. The paper, titled Reinventing universities for today's world, explores the transformative potential of artificial intelligence in higher education, with a particular focus on accelerating medical training through AI-driven innovation. In the face of escalating global challenges such as climate change and public health emergencies, higher education institutions are under growing pressure to modernize and improve the efficiency of talent development. Zhang argues that while generative AI tools like ChatGPT pose challenges to traditional assessment models, they should be embraced as revolutionary accelerators of learning rather than threats. Zhang highlighted research led by Professor Yang Liu from AIR, which developed an AI-powered simulation system called "Agent Hospital." This system uses multimodal large models to create a fully immersive virtual medical environment, simulating the entire patient journey—from initial triage and registration, to medical examinations, diagnosis, treatment planning, and follow-up care. The platform features AI-driven virtual doctors and patients, enabling trainees to engage in realistic clinical scenarios. According to Zhang, the AI doctors are trained on millions of real-world medical cases, effectively replicating decades of human clinical experience. In benchmark tests using the MedQA dataset—based on the U.S. Medical Licensing Examination—the AI system achieved a 96% diagnostic accuracy rate for both virtual and real patient symptoms, marking a promising milestone. The system is currently being piloted at Tsinghua University to train future physicians. Virtual patients communicate symptoms through text, speech, and simulated physical movements. Researchers are also developing wearable devices equipped with haptic sensors to replicate tactile sensations, such as skin texture and temperature, further enhancing realism. Zhang emphasized that while AI cannot replace hands-on training with real patients under physician supervision, it offers immense value in early-stage education. "Using these tools at the beginning of training allows students to rapidly acquire knowledge and skills in a safe, risk-free environment—without the ethical or safety concerns associated with real-world patient interaction," he said. The article also discusses broader applications of AI simulation across higher education, including administrative tasks, literature review, and coding. However, the authors stress the importance of ethical oversight, warning that AI tools must support, not replace, human judgment. Agent Hospital exemplifies China’s growing leadership in the integration of AI and education. As such technologies scale globally, they hold the potential to overcome long-standing bottlenecks in high-level talent development, driving higher education toward greater efficiency, accessibility, and inclusivity. Nature, Vol. 645, 25 September 2025, pp. 852–855.

Related Links