2026 ASEE Annual Conference & Exposition

Evaluating a ChatGPT-Based Hyponatremia Teaching Module for Internal Medicine Residents: A Pilot Study

Presented at Generative AI in BME Courses

Generative artificial intelligence (GenAI) is increasingly used in medical education to enhance engagement and simplify complex topics. Nephrology, often perceived as challenging by trainees, may particularly benefit from such innovations. This pilot study evaluated whether a ChatGPT-based interactive hyponatremia learning module, enhanced with Retrieval-Augmented Generation (RAG) using content from Pocket Nephrology, could support internal medicine resident education. Residents were randomized 2:1 to intervention (module + quiz) or control (quiz only, open-resource). Of 28 participants (19 intervention, 9 control), mean quiz scores were similar (7.47 ±1.90 vs. 7.44 ± 1.94; p = 0.97). However, intervention group confidence increased from 2.67 ±1.23 to 3.50 ± 0.80 on a 5-point scale. Usability ratings were high for clarity and appropriateness. Expert review of 14 transcripts found most outputs accurate, though occasional errors persisted despite RAG. A ChatGPT-based, RAG-enhanced module was feasible, well-received, and improved resident confidence, though it did not significantly increase short-term quiz performance. Integration of GenAI in nephrology education shows promise but requires expert oversight.

Authors

Dr. Christelle El Khoury Lebanese American University [biography]

Christelle El Khoury, MD, is a family physician with a Master's in Applied AI from the Lebanese American University (LAU). She previously worked with the Department of Family Medicine at the University of Michigan Medical School on research focused on cervical cancer screening and patient-centered exam alternatives. She has also worked as a medical expert in creating clinical prompts to train large language models for medical and educational applications. Her current interests include leveraging AI to enhance medical education and strengthen patient engagement.
Dr. Toni Sabbouh MD, FASN University of Pennsylvania [biography]

Toni Sabbouh, MD, FASN, is an American board‑certified physician in Internal Medicine and Nephrology and a fellowship‑trained nephrologist from the University of Michigan Hospitals. A Fellow of the American Society of Nephrology, he is recognized for excellence in clinical care and scholarly contributions. Dr. Sabbouh currently serves as a Clinical Assistant Professor of Medicine at Penn Medicine in Philadelphia, specializing in kidney disease and advanced clinical management.
Dr. Christine Palazzolo DO University of Pennsylvania [biography]

Christine Palazzolo, DO, is a physician who completed her Internal Medicine residency and served as Chief Resident at Pennsylvania Hospital. She will begin fellowship training in Nephrology at the University of Pennsylvania in 2026. She is actively involved in patient advocacy and women in medicine initiatives and has published with the American Society of Nephrology. Her academic interests include palliative nephrology and improving patient-centered outcomes in kidney disease.
Dr. Sola Aoun Bahous Lebanese American University [biography]

Dr. Aoun Bahous is Professor of Medicine and Dean of the Lebanese American University School of Medicine. A nephrologist, pharmacologist, and medical educator, she has led innovations in medical curricula, including the early integration of Artificial Intelligence and Digital Health across all medical years since 2018. Her research interests include hypertension, electrolyte disorders, and medical education. She serves on the International Advisory Committee on Artificial Intelligence in Medical Education and as Deputy Chair of the International Society of Nephrology’s Education Working Group. With a particular interest in pedagogy, she focuses on leveraging innovative educational tools to make complex nephrology concepts more accessible and engaging for learners.
Dr. Haidar Harmanani http://orcid.org/0000-0001-5416-4383 Lebanese American University [biography]

Haidar Harmanani received the B.S., M.S. and Ph.D. degrees in Computer Engineering from Case Western Reserve University, Cleveland, Ohio, in 1989, 1991 and 1994, respectively. He is currently a professor of computer science and the Dean of the School of Arts and Sciences at the Lebanese American University. His research interests span high-level synthesis, design automation, parallel computing, machine learning, and applied artificial intelligence. Throughout his career, he has led and contributed to multiple institutional initiatives in academic program development, accreditation, research infrastructure, and strategic planning. Dr. Harmanani has published extensively in the areas of design automation and optimization techniques for hardware–software systems, and he has supervised numerous graduate and undergraduate research projects in computing and data-driven systems.

Download paper (364 KB)

Are you a researcher? Would you like to cite this paper? Visit the ASEE document repository at peer.asee.org for more tools and easy citations.

» Download paper

« View session