Research
We focus on how technology could be used to enhance teaching and learning experience.
One of our current directions is the development and training of medical large language model (M-LLM). This is different from daily conversation or online discussion in various aspects such as the number of utterance in a conversation, the number of tokens in an utterance, etc.
Our goal is to develop LLMs for training different specialists and professionals to interactive with the people.
Past Research
- GenAI Medical Education
-
We studied how generative AI can be used to train medical students
-
We provided theoretical support for the effectiveness of using chatbots for medical education
Using clinical history taking chatbot mobile app for clinical bedside teachings - A prospective case control study.
Heliyon. Volume 8, Issue 6, June 2022.
Abstract: A novel chatbot mobile app for training of undergraduate medical students' clinical history taking skills was developed in 2021. Students were able to take clinical history from the virtual patient for bedside teaching. A case-control study was conducted to evaluate the effectiveness of learning with chatbot mobile app, versus conventional bedside teachings with real patients. With the promising results we have demonstrated in this study, training of history taking skills by chatbot will be a feasible alternative to conventional bedside teaching.
ChatGPT versus human in generating medical graduate exam multiple choice questions-A multinational prospective study (Hong Kong S.A.R., Singapore, Ireland, and the United Kingdom).
PLoS One. 2023 Aug 29;18(8).
Abstract: Given the substantial workload of university medical staff, this study aims to assess the quality of multiple-choice questions (MCQs) produced by ChatGPT for use in graduate medical examinations, compared to questions written by university professoriate staffs based on standard medical textbooks. ChatGPT has the potential to generate comparable-quality MCQs for medical graduate examinations within a significantly shorter time.
Large language models: implications of rapid evolution in medicine.
Hong Kong Med J 2023;29:Epub 11 Dec 2023.
Abstract: We are at a pivotal juncture in history where artificial intelligence and LLMs are becoming increasingly reliable and usable tools with the potential to revolutionise clinical practice, scientific discovery, and teaching. We encourage clinicians and researchers to harness the power of these tools, think creatively about their applications, and continually explore ways in which they can enhance our work, drive scientific advancements, and ultimately benefit humanity as a whole.
Using ChatGPT for medical education: the technical perspective.
BMC Medical Education (2025) 25:201, 2025.
Abstract: The objective of this research is to examine the influence of the newly proposed chatbot on learning efficacy and experiences in bedside teaching, and its potential contributions to international teaching collaboration. This study employs a mixed-method design that incorporates both quantitative and qualitative approaches. From the quantitative approach, we launched the world’s first cross-territory virtual bedside teaching with our proposed application and conducted a survey between the University of Hong Kong (HKU) and the National University of Singapore (NUS). Descriptive statistics and Spearman’s Correlation were applied for data analysis. From the qualitative approach, a comparative analysis was conducted between the two versions of the chatbot. And, we discuss the inter-relationship between the quantitative and qualitative results..
Past Research
- Medicine, Education & Technology
-
We studied how technologies such as virtual reality and virtual classroom can be used to train medical students
Extended reality in surgical education: A systematic review.
Surgery. 2023 Nov; 174(5):1175-1183
Abstract: Evaluate the effectiveness of extended reality-based training in surgical education. Extended reality-based training is a potentially useful modality to serve as an adjunct to the current physical surgical training.
Distance education for anatomy and surgical training – A systematic review.
The Surgeon. Volume 20, Issue 5, October 2022, 195-205.
Abstract: Rapid development of COVID-19 has resulted in a massive shift from traditional to online teaching. This review aims to evaluate the effectiveness of distance learning on anatomy and surgical training. In conclusion, distance learning is a feasible alternative for anatomy and surgical teaching.
Past Research
- Medical Privacy
-
We studied how privacy enhancing technologies can be used to protect medical data
-
Strong protection in privacy ensures the availability of data for training AI for medical research
Efficient unbalanced private set intersection cardinality and user-friendly privacy-preserving contact tracing.
USENIX 2023.
Abstract: An unbalanced private set intersection cardinality (PSI-CA) protocol is a protocol to securely get the intersection cardinality of two sets X and Y without disclosing anything else, in which |Y| < |X|. We apply our lightweight unbalanced PSI-CA protocols to design a privacy-preserving contact tracing system. Our system outperforms existing schemes in terms of security and performance.
Privacy-preserving COVID-19 contact tracing app: A zero-knowledge proof approach.
ISPEC 2021.
Abstract: We propose a privacy-preserving contact tracing protocol for Android and iOS phones. The protocol allows users to be notified, if they have been a close contact of a confirmed patient. The app allows all users to hide their past location(s) and contact history from the Government, without affecting their ability to determine whether they have close contact with a confirmed patient whose identity will not be revealed. A zero-knowledge protocol is used to achieve such a user privacy functionality.
A general framework for secure sharing of personal health records in cloud system.
J. Comput. Syst. Sci. 2017
Abstract: Personal Health Record (PHR) has been developed as a promising solution that allows patient–doctors interactions in a very effective way. We propose a general framework for secure sharing of PHRs. Our system enables patients to securely store and share their PHR in the cloud server (for example, to their carers), and furthermore the treating doctors can refer the patients' medical record to specialists for research purposes, whenever they are required, while ensuring that the patients' information remain private.