Researchers at Weill Cornell Medicine have developed an artificial intelligence system called EmulatRx that could significantly speed up the design of clinical trials by using real-world patient data. Published in Nature Communications, the system is designed to simplify one of the most complex stages of drug development by helping researchers determine eligibility criteria, treatment strategies, study endpoints, and patient recruitment plans more efficiently.
Unlike a single AI model, EmulatRx operates as a virtual team of five specialized AI agents, each performing a different role typically handled by human experts. The system includes a Supervisor to coordinate the workflow, a Trialist to analyze previous studies, an Informatician to identify suitable patients from electronic health records, a Clinician to validate medical decisions, and a Statistician to evaluate likely trial outcomes. Together, these agents collaborate to refine clinical trial designs using natural language and real-world healthcare data.
The researchers tested EmulatRx using electronic health records and target trial emulation techniques, enabling it to simulate randomized clinical trials before they are conducted. By combining structured medical data with clinical notes, the system can identify eligible patient groups, estimate treatment effectiveness, and improve study designs while reducing the time and cost involved in planning traditional clinical trials.
Although the technology is still in the research stage and requires broader validation across different healthcare systems, the team believes it has the potential to make clinical trials faster, more affordable, and more precise. By improving trial design before studies begin, AI systems like EmulatRx could accelerate the development of new medicines and increase the likelihood of successful treatments reaching patients sooner.
