This research introduces a user-centered VR simulation system that leverages LLMs to create realistic virtual patient interactions for healthcare professionals in training.

• Addresses current limitations of fixed content in VR medical simulations
• Employs LLM-based embodied conversational agents for dynamic, customizable patient scenarios
• Developed through interviews with advanced health profession students to identify communication challenges
• Creates a Virtual AI Patient Simulator enabling practice in a safe, flexible environment

This innovation matters for medical education by providing scalable, consistent training opportunities outside clinical settings, helping future healthcare professionals develop crucial communication skills without patient risk.

Designing VR Simulation System for Clinical Communication Training with LLMs-Based Embodied Conversational Agents