Use of a Large Language Model as a Dermatology Case Narrator: Exploring the Dynamics of a Chatbot as an Educational Tool in Dermatology
JMIR DermatologyResearch Authors: Emmanouil Karampinis, MSc, MD, PhD; Dafni Anastasia Bozi Tzetzi, MSc, MD; Georgia Pappa, MD; Dimitra Koumaki, MSc, MD, PhD; Dimitrios Sgouros, MD, PhD; Efstratios Vakirlis, MD, PhD; Aikaterini Liakou, MD, PhD; Markos Papakonstantis, MD; Marios Papadakis, MD, PhD; Dimitrios Mantzaris, MSc, PhD; Elizabeth Lazaridou, MD, PhD; Enzo Errichetti, MSc, MD; Cristián Navarrete-Dechent, MD; Angeliki Victoria Roussaki Schulze, MD, PhD; Alexandros Katoulis, MD, PhDAIIM Authors: Hanna Zhu, Josh BronteApproved by President Reda RiffiPublication Date: 9/16/2025Comprehensive Summary
This study examines how large language models, specifically ChatGPT-4, can be used to create dermatology case-based scenarios for medical students. 64 medical students were given both AI and non-AI generated cases on similar topics to evaluate the educational, motivational, and challenging aspects of each case. Each question was graded on the Likert scale from 1-5. The results showed that although the AI-generated cases were more detailed in medical history and descriptions, they lacked the clinical depth and motivational elements of non-AI generated cases. Students rated the non-AI generated cases to be more educational, engaging, and challenging throughout different conditions such as psoriasis, dermatitis, rosacea, etc. The authors emphasize that AI chatbots can be used as supplemental tools, but should not be relied on completely.
Outcomes and Implications
Understanding the strengths and weaknesses of AI-generated case scenarios is important in how medical schools explore new technologies to support learning, which heavily relies on case-based learning. This study highlights that AI can help in the creation of educational materials, but should not be used as a replacement. Future improvements of AI models may help the application of its use, but non-AI cases are still necessary for learning.
Connect medicine with AI innovation.
No spam. Only the latest AI breakthroughs, simplified and relevant to your field.