Exploring the Potential and Limitations of Chat Generative Pre-trained Transformer (ChatGPT) in Generating Board-Style Dermatology Questions: A Qualitative Analysis
Open accessThe DOI matches its Crossref record; 5 of the 5 references it cites were found in a registry or an index.
Abstract
This article investigates the limitations of Chat Generative Pre-trained Transformer (ChatGPT), a language model developed by OpenAI, as a study tool in dermatology. The study utilized ChatPDF, an application that integrates PDF files with ChatGPT, to generate American Board of Dermatology Applied Exam (ABD-AE)-style questions from continuing medical education articles from the Journal of the American Board of Dermatology. A qualitative analysis of the questions was conducted by two board-certified dermatologists, assessing accuracy, complexity, and clarity. Out of 40 questions generated, only 16 (40%) were deemed accurate and appropriate for ABD-AE study preparation. The remaining questions exhibited limitations, including low complexity, lack of clarity, and inaccuracies. The findings highlight the challenges faced by ChatGPT in understanding the domain-specific knowledge required in dermatology. Moreover, the model's inability to comprehend the context and generate high-quality distractor options, as well as the absence of image generation capabilities, further hinders its usefulness. The study emphasizes that while ChatGPT may aid in generating simple questions, it cannot replace the expertise of dermatologists and medical educators in developing high-quality, board-style questions that effectively evaluate candidates' knowledge and reasoning abilities.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Kung, T. H., Cheatham, M., Medenilla, A., Sillos, C., De Leon, L., Elepaño, C., Madriaga, M., Aggabao, R., Diaz-Candido, G., Maningo, J., et al. (2023) · PLOS Digital Health · doi:10.1371/journal.pdig.0000198 · Cited by 4,115 · cited 1×
“Furthermore, ChatGPT has been employed for taking licensing examinations and responding to specialty board review questions, demonstrating an average accuracy rate close to passing thresholds [6,7].”
MethodKung, T. H., Cheatham, M., ChatGPT, Medenilla, A., Sillos, C., De Leon, L., Elepaño, C., Madriaga, M., Aggabao, R., Diaz-Candido, G., et al. (2022) · doi:10.1101/2022.12.19.22283643 · Cited by 4,115
Eysenbach, G. (2023) · JMIR Medical Education · doi:10.2196/46885 · Cited by 840 · cited 1×
“Chat Generative Pre-trained Transformer (ChatGPT) is a language model developed by OpenAI (San Francisco, CA, USA) that has shown promise in various natural language processing (NLP) tasks, including medical education and multiple-choice question generation [1,2].”
MethodThe rise of ChatGPT : Exploring its potential in medical education
Lee, H. (2024) · Anatomical Sciences Education · doi:10.1002/ase.2270 · Cited by 596 · cited 1×
“Chat Generative Pre-trained Transformer (ChatGPT) is a language model developed by OpenAI (San Francisco, CA, USA) that has shown promise in various natural language processing (NLP) tasks, including medical education and multiple-choice question generation [1,2].”
MethodChatGPT for healthcare providers and patients: Practical implications within dermatology
Jin, J. Q., & Dobry, A. S. (2023) · Journal of the American Academy of Dermatology · doi:10.1016/j.jaad.2023.05.081 · Cited by 49
Cited by
A scoping review of artificial intelligence in medical education: BEME Guide No. 84
Gordon, M., Daniel, M., Ajiboye, A., Uraiby, H., Xu, N. Y., Bartlett, R., Hanson, J., Haas, M., Spadafore, M., Grafton-Clarke, C., et al. (2024) · Medical Teacher · doi:10.1080/0142159x.2024.2314198 · Cited by 679
Tam, T. Y. C., Sivarajkumar, S., Kapoor, S., Stolyar, A. V., Polanska, K., McCarthy, K. R., Osterhoudt, H., Wu, X., Visweswaran, S., Fu, S., et al. (2024) · npj Digital Medicine · doi:10.1038/s41746-024-01258-7 · Cited by 424
Large language models for generating medical examinations: systematic review
Artsi, Y., Sorin, V., Konen, E., Glicksberg, B. S., Nadkarni, G., & Klang, E. (2024) · BMC Medical Education · doi:10.1186/s12909-024-05239-y · Cited by 128
“Ayub et al. [33] focused on medical board examination for Dermatology.”
BackgroundMorjaria, L., Burns, L., Bracken, K., Levinson, A. J., Ngo, Q. N., Lee, M., & Sibbald, M. (2024) · International Medical Education · doi:10.3390/ime3010004 · Cited by 33
“…body of research that addresses ChatGPT performance on answering short-answer exam questions [22–24] and its ability to generate exam problems [15,25,26], there are presently no published studies that evaluate the ability of ChatGPT to score medical student answers to short-answer exam questions.”
Metadata
- Authors
- Ayub, I., Hamann, D., Hamann, C. R., & Davis, M. J.
- Venue
- Cureus
- Published
- 2023-08-18
- Volume
- 15
- Issue
- 8
- Pages
- e43717-e43717
- ISSN
- 2168-8184 (ISSN-L)
- OpenAlex
- W4385972264
- Cited by
- 53
- References
- 5
- FWCI
- 1.83
- Type
- journal article
- Language
- en
- Open access
- diamond
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.