Conversational Robots as Virtual Patients

Can robots simulate patients and be used as a tool to train medical students on how to talk to patients and develop their clinical reasoning skills?

This project focuses on developing an advanced virtual patient (VP) platform utilizing social robots and large language models (LLMs) to enhance clinical teaching in rheumatology for medical students at the Karolinska Institute (KI), Stockholm. While VPs are already used in medical education, current platforms lack the interactivity and realism of real-life patient encounters. Our platform, built on the social robot Furhat, aims to improve clinical reasoning skills by enabling more natural and interactive dialogue with virtual patients, in order to simulate realistic doctor-patient interactions.

Publications

[1]

A. Borg, I. Parodis and G. Skantze, "Creating Virtual Patients using Robots and Large Language Models : A Preliminary Study with Medical Students," in HRI 2024 Companion - Companion of the 2024 ACM/IEEE International Conference on Human-Robot Interaction, 2024, pp. 273-277.

[2]

A. Borg et al., "Enhancing clinical reasoning skills for medical students : a qualitative comparison of LLM-powered social robotic versus computer-based virtual patients within rheumatology," Rheumatology International, 2024.