Meet one of our Research Grant submissions: Investigating the application of artificial intelligence (AI) driven virtual patients (VPs) for healthcare students to develop communication skills

As a lecturer and PhD candidate from the Department of Audiology and Speech Pathology, my research aims to build the evidence base around AI-driven virtual patients (VPs) for enhancing health professions students' communication skills. This project originated in 2022, with initial planning and groundwork preceding the launch of ChatGPT in November 2022. The surge of interest and possibilities in this field since then has been truly remarkable.

Alongside my supervisors, I am exploring this area due to the challenges posed by increasing student numbers, limited placement opportunities, and the well-established importance of effective patient-clinician communication. Our goal is to investigate the landscape of conversational AI agents representing patients and consult key stakeholders about their attitudes and needs regarding these emerging learning tools. Following from this, we aim to investigate if and how these tools may reach their goal of furthering students’ development of communication skills.

Through a scoping review published earlier this year (https://doi.org/10.14742/ajet.9307), we have examined the tools discussed in the literature. Subsequently, we have focused on understanding the attitudes of students and educators towards these tools. It is crucial to carefully consider the pedagogical approach when incorporating AI-driven VPs into our curricula. The insights from these stakeholder groups have been invaluable in identifying design and implementation factors to consider for the development and use of AI-driven VPs. We hope to share these findings broadly in the near future.

Moving forward, we are concentrating on investigating solutions that continue to incorporate student voices through co-design and address the needs identified through our consultations with educators and students to date. It is essential to find cost-effective solutions for all healthcare programs that can effectively target the desired learning objectives.

If you have an interest in this project or the broader area of research and teaching involving AI-driven VPs, please reach out. I would love to discuss it further!

- Patrick Bowers, University of Melbourne.