ArticleJournal of cosmetic dermatology2025
The Performance of AI in Dermatology Exams: The Exam Success and Limits of ChatGPT.
Article in Journal of cosmetic dermatology, 2025. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Cited by 2 papers.
What it found
Each row is one number read from the abstract, on the scale the paper reported it, with its interval. Left of the dashed line favours the treatment, right favours the comparator. Under each row is the sentence it came from. New to these charts? A ten-minute tutorial.
The abstract states no effect estimate the extractor could read, or names no intervention and outcome on the map, so this paper lights no cell and moves no belief. It is still indexed, cited and linked below.
The trial behind it
Trials whose registry record cites this paper, or whose number appears in the abstract. A trial that started after this paper was published is citing it as background, not reporting it.
Neither the registry nor the abstract names a trial number. If this is a trial report, that itself is worth knowing.
Who cites it
2 citing papers in PubMed.
- Artificial intelligence in dermatology: A literature review of current evidence and clinical implementation.JID innovations : skin science from molecules to population health · 2026Review
- Humans vs. large language models in neurology board examination: performance, limitations, and reference reliability.Acta neurologica Belgica · 2026Article
Corrections and comments
PubMed lists nothing against this paper. Absence here is not a guarantee, only a check that was made.
Authors and funding
2 authors.
Funding
No grant is acknowledged in the PubMed record.
Abstract
backgroundArtificial intelligence holds significant potential in dermatology.
objectivesThis study aimed to explore the potential and limitations of artificial intelligence applications in dermatology education by evaluating ChatGPT's performance on questions from the dermatology residency exam.
methodIn this study, the dermatology residency exam results for ChatGPT versions 3.5 and 4.0 were compared with those of resident doctors across various seniority levels. Dermatology resident doctors were categorized into four seniority levels based on their education, and a total of 100 questions-25 multiple-choice questions for each seniority level-were included in the exam. The same questions were also administered to ChatGPT versions 3.5 and 4.0, and the scores were analyzed statistically.
resultsChatGPT 3.5 performed poorly, especially when compared to senior residents. Second (p = 0.038), third (p = 0.041), and fourth-year senior resident physicians (p = 0.020) scored significantly higher than ChatGPT 3.5. ChatGPT 4.0 showed similar performance compared to first- and third-year senior resident physicians, but performed worse in comparison to second (p = 0.037) and fourth-year senior resident physicians (p = 0.029). Both versions scored lower as seniority and exam difficulty increased. ChatGPT 3.5 passed the first and second-year exams but failed the third and fourth-year exams. ChatGPT 4.0 passed the first, second, and third-year exams but failed the fourth-year exam. These findings suggest that ChatGPT was not on par with senior resident physicians, particularly on topics requiring advanced knowledge; however, version 4.0 proved to be more effective than version 3.5.
conclusionIn the future, as ChatGPT's language support and knowledge of medicine improve, it can be used more effectively in educational processes.
Indexed as
Identifiers
What OpenQuestion holds
Registered trials
Read under generation 80e0d062 · epoch 390. Bibliography from PubMed, PubMed Central and OpenAlex; grants from NIH RePORTER; trial links from ClinicalTrials.gov; estimates, votes and beliefs from the OpenQuestion graph.