Related Experiment Video
Updated: Jun 6, 2026
07:14
Virtual Agent for Real-Time Motivational Interviewing by Integrating Adaptive Nonverbal Behavior and Language Models
Published on: December 23, 2025
A Large Language Model-Powered Multiagent Framework Emulating Standardized Patients in Clinical Communication Skills
Yufei Qu1, Xiaowei Xu1,2, Yunzi Long3,4
1College of Biomedical Engineering and Instrument Science, Zhejiang University, No. 38 Zheda Road, Hangzhou, 310058, China.
Journal of Medical Internet Research
|June 4, 2026
Summary
A new multiagent virtual patient framework effectively simulates standardized patients for medical training, outperforming single large language models in role-playing and interaction fidelity. This approach enhances clinical communication skills for students.
Area of Science:
- Artificial Intelligence in Medical Education
- Computational Linguistics for Healthcare Simulation
- Multiagent Systems in Clinical Training
Background:
- Effective clinical communication is vital in medical practice.
- Standardized patients (SPs) are a reliable training method but resource-intensive.
- Current virtual patients (VPs) using single large language models (LLMs) face fidelity and interaction limitations.
Purpose of the Study:
- To develop and evaluate a novel multiagent VP framework simulating SPs.
- To enhance human-like fidelity and interaction performance in clinical communication training.
- To leverage collaborative agent design for improved VP simulation.
Main Methods:
- Constructed a multiagent framework with 5 specialized subagents simulating brain region functions.
- Incorporated retrieval-augmented technology and deep character reasoning for enhanced interaction.
- Evaluated the framework by comparing base models and benchmarking against single-LLM baselines using metrics like response quality and role-playing performance.
Main Results:
- The multiagent framework surpassed single-LLM baselines in accuracy and role-playing under standardized conditions.
- GPT-4o implementation achieved 0.769 factual consistency; all configurations maintained >94% clinical accuracy.
- Qwen3-32B framework showed a low misleading rate (1.28%) and high role-playing competency (39.67), with practical interaction latency (~3s).
Conclusions:
- The multiagent framework provides a viable, customizable, and scalable simulation of SPs for medical communication training.
- This approach enhances VP performance, maintaining patient confidentiality.
- The framework shows significant potential for advancing medical education methodologies.
Related Concept Videos
01:26Modeling in Therapy
Modeling, a key technique in therapy, uses observational learning to help clients acquire and practice new skills by watching therapists demonstrate desired behaviors. This approach, rooted in Albert Bandura's concept of vicarious learning, plays a significant role in therapeutic interventions for various psychological conditions, including social anxiety, ADHD, and depression.
Participant Modeling
Participant modeling involves therapists demonstrating calm and effective behaviors in situations...
Participant Modeling
Participant modeling involves therapists demonstrating calm and effective behaviors in situations...
01:14SBAR II: Application of SBAR
SBAR is an effective communication tool used by healthcare professionals to communicate patient information accurately. SBAR stands for Situation, Background, Assessment, and Recommendation. For a better understanding, an example is given below.
SBAR Report from a Nurse to a Health Care Provider
S: "Hello, Dr. Smith. This is Jane, RN, from the Med Surg unit. I am calling to tell you about Ms. White in Room 210, who is experiencing increased pain and redness at her incision site. Her recent...
SBAR Report from a Nurse to a Health Care Provider
S: "Hello, Dr. Smith. This is Jane, RN, from the Med Surg unit. I am calling to tell you about Ms. White in Room 210, who is experiencing increased pain and redness at her incision site. Her recent...