Abstract
Effective physician-patient communication is fundamental to clinical competence, yet traditional simulation-based training methods using standardized patients and high-fidelity manikins are costly, resource-intensive, and difficult to scale. This dissertation presents CLiVR (Conversational Learning system in Virtual Reality), an LLM-driven system that integrates large language models and 3D avatars to simulate doctor-patient interactions for medical communication training.
CLiVR addresses three key limitations in existing virtual reality medical training platforms. First, the system operates on standalone Meta Quest 3 hardware with realistic 3D patient avatars featuring synchronized lip movements and speech-based interaction. Second, CLiVR grounds LLM responses using a curated syndrome-symptom database, constraining patient dialogue to clinically accurate scenarios while preserving conversational flexibility. Third, the system employs sentiment analysis to classify trainee communication tone, providing quantitative feedback on empathy and interpersonal effectiveness.
The system was evaluated through a user study with medical school faculty (n=13). Quantitative results indicated acceptance of the technology for teaching physician-patient interactions. Qualitative feedback identified the platform's value in providing repeatable practice opportunities while noting areas for enhancement including expanded patient diversity and integration of clinical data. These findings demonstrate the feasibility of LLM-driven VR systems as scalable supplements to traditional standardized patient training.
Date of publication
2026
Document Type
Thesis
Language
english
Persistent identifier
http://hdl.handle.net/10950/5046
Committee members
Sagnik Dakshit, Arun Kulkarni, Kouider Mokhtari
Degree
Master of Science in Computer Science
Recommended Citation
Amithasagaran, Akilan, "AN LLM-DRIVEN SYSTEM FOR DOCTOR-PATIENT SIMULATION" (2026). Computer Science Theses. Paper 7.
http://hdl.handle.net/10950/5046