Abstract
As telemedicine becomes a global standard of care, equipping future clinicians with digital consultation skills is critical. Yet in Hong Kong, telemedicine education remains fragmented, lacking an integrated, contextually relevant curriculum. This Teaching Development Grant (TDG) project aims to fill that gap by piloting a longitudinal, outcomes-based telemedicine curriculum within the MBBS Family Medicine and Primary Care (FMPC) programme at the University of Hong Kong.
A key innovation of this curriculum is the use of generative artificial intelligence (GenAI), specifically large language models like ChatGPT, to support scalable, personalized feedback after student-led teleconsultation exercises. Students in clinical years conduct virtual consultations with standardized patients. Transcripts of these encounters are analysed by GenAI to generate formative feedback on clinical reasoning, communication, empathy, and documentation. A clinical tutor then reviews and validates this AI-generated feedback before delivering it to students. This hybrid model leverages GenAI to overcome common educational barriers—such as limited tutor bandwidth—while preserving feedback quality and safety.
The broader curriculum also includes two interactive E-Modules and ethics-focused workshops alongside teleconsultation practicum sessions. Students learn key telehealth competencies, including “webside manner”, legal and ethical considerations, and disposition planning. The curriculum is aligned with the international telehealth competency frameworks and contextualized for Hong Kong’s healthcare and legal landscape.
The study is ongoing, with evaluation guided by Kirkpatrick’s four-level model. Pre- and post-tests, usability questionnaires, Tele-OSCEs, and stakeholder interviews will measure knowledge gains, skill acquisition, and educational value. Preliminary student feedback highlights the GenAI feedback loop as a timely, individualized learning aid that complements live supervision.
This project demonstrates how thoughtfully applied GenAI can enhance feedback in complex, communication-heavy domains like telemedicine. It holds promise not only for medical education but also for interdisciplinary domains such as Law, Business, and Social Work where virtual consultations and ethical communication are vital. As AI continues to reshape education, our findings will inform best practices for integrating GenAI into feedback systems in a safe, ethical, and learner-centered manner.
Presenter(s) / Author(s)
