AI in Simulated Patients

•  Body Interact - Virtual Patients Simulator  
•  Virtual Reality Medical Simulation | SimX 
•  Digital Clinical Experiences™ | Shadow Health® 

AI-powered VSPs simulate realistic patient interactions using natural language processing (NLP) and machine learning:

• Functionality: Learners can speak or type to these virtual patients and receive realistic responses. 
• Educational Impact: Helps students practice communication, diagnosis, and clinical reasoning in a safe environment. 
• Examples: 
  • SimX and OSSimTech integrate AI avatars in virtual reality simulations. 
  • Shadow Health (now part of Elsevier) uses conversational AI in digital clinical experiences.

• Ethical Use: Ensuring equity, privacy, and transparency in AI-driven feedback. 
• Fidelity: While improving, AI-generated emotional responses and nonverbal cues may still lag behind human actors. 
• Faculty Training: Instructors must learn how to integrate and interpret AI tools effectively.

The University of California, San Francisco (UCSF) developed MedSimAI, an AI-powered simulation platform that enables deliberate practice and self-regulated learning through interactive patient encounters. Leveraging large language models (LLMs), MedSimAI generates realistic clinical interactions and provides immediate, structured feedback using established medical evaluation frameworks. In a pilot study with 104 first-year medical students, participants found MedSimAI beneficial for repeated, realistic patient-history practice. The platform addresses key limitations of traditional simulation-based training, making high-quality clinical education more accessible and scalable. arXiv

A study in Japan investigated whether medical students' interview skills could be improved by engaging with AI-simulated patients using large language models. The simulation program provided to fourth-year medical students demonstrated that AI-driven virtual patients can effectively enhance medical interview skills, offering a scalable and consistent training tool. JMIR Medical Education+1PMC+1

Ohio State University developed a Virtual Standardized Patient (VSP) system that allows students to practice history-taking skills by interacting with AI-powered avatars. These virtual patients, created in Unity and controlled by ChatScript, enable students to conduct medical histories and develop differential diagnoses, providing immediate feedback and a safe environment for skill development. U.OSU+1Cureus+1

Virti, a platform utilizing VR and AR technologies, introduced AI-powered virtual patients to enhance remote clinical training. During the COVID-19 pandemic, Virti's technology was instrumental in training healthcare workers across the NHS and hospitals in the US. The platform provides detailed feedback and metrics on trainee performance, helping identify areas for improvement. Wikipedia

A randomized controlled trial published in BMC Medical Education compared AI-based simulators with traditional simulated patients in teaching history-taking skills to undergraduate medical students. The study aims to evaluate the effectiveness of AI simulators in delivering simulation sessions and their impact on learning outcomes, student satisfaction, and engagement. Findings from such studies will guide decisions regarding integrating AI-based simulators into healthcare education and training programs. BioMed Central+1U.OSU+1
 

• NEJM AI
• Teaching Robots How to Act Like Humans