Chinese researchers from Tsinghua University have developed “Agent Hospital,” an AI-powered virtual hospital town similar to Stanford’s AI town, which gained widespread attention last year. The innovative system features AI doctors and nurses autonomously treating virtual patients, with the goal of improving medical expertise through continuous evolution.
The AI agents, powered by large language models (LLMs), autonomously interact within the virtual hospital environment. According to the researchers, the AI doctors can treat up to 10,000 patients in a few days, a task that would take human doctors at least two years. These AI doctors have demonstrated impressive accuracy, achieving a 93.06% success rate on the MedQA dataset, which involves the full process of patient care, from diagnosis to follow-up.
Agent Hospital includes various consultation and examination rooms staffed by 14 AI doctors and four nursing agents, allowing for risk-free medical training. This setup is designed not only to optimize operations but also to help medical students practice developing treatment plans without real-world consequences.
Liu Yang, leader of the project, emphasized the potential for AI hospital towns to transform healthcare. “This concept holds immense significance for both medical professionals and the general public,” he told the Beijing-based Global Times. The virtual hospital aims to continuously train and improve AI doctors, offering high-quality, affordable, and convenient healthcare services. It also allows for simulations of real-world medical scenarios, such as predicting the spread of infectious diseases.
Despite the promising potential of AI-driven healthcare, challenges remain, particularly in adhering to national medical regulations and ensuring effective AI-human collaboration. Dr. Dong Jiahong of Tsinghua University stressed that while AI can boost healthcare efficiency and precision, it cannot replace the personalized care and compassion provided by human doctors.
By Impact Lab