MétaCan
Menu
Back to cohort
Record W4406336691 · doi:10.1111/ddg.15609

The use of a ChatGPT‐4‐based chatbot in teledermatology: A retrospective exploratory study

2025· article· en· W4406336691 on OpenAlexaff
Jonathan Shapiro, Emily Avitan‐Hersh, Binyamin Greenfield, Ziad Khamaysi, Roni P. Dodiuk‐Gad, Yuliya Valdman‐Grinshpoun, Tamar Freud, Anna Lyakhovitsky

Bibliographic record

VenueJDDG Journal der Deutschen Dermatologischen Gesellschaft · 2025
Typearticle
Languageen
FieldMedicine
TopicArtificial Intelligence in Healthcare and Education
Canadian institutionsUniversity of Toronto
Fundersnot available
KeywordsTeledermatologyConcordanceMedical diagnosisMedicineWorkflowMedical physicsComputer scienceArtificial intelligenceTelemedicineRadiologyHealth careInternal medicineDatabase

Abstract

fetched live from OpenAlex

BACKGROUND AND OBJECTIVES: Integration of artificial intelligence in healthcare, particularly ChatGPT, is transforming medical diagnostics and may benefit teledermatology. This exploratory study compared image description and differential diagnosis generation by a ChatGPT-4 based chatbot with human teledermatologists. PATIENTS AND METHODS: This retrospective study compared 154 teledermatology consultations (December 2023-February 2024) with ChatGPT-4's performance in image descriptions and diagnoses. Diagnostic concordance was classified as "Top1" (exact match with the teledermatologist's diagnoses), "Top3" (correct diagnosis within one the top three diagnoses), and "Partial" (similar but not identical diagnoses). Image descriptions were rated and compared for quality parameters (location, color, size, morphology, and surrounding area), and accuracy (Yes, No, and Partial). RESULTS: Out of 154 cases, ChatGPT-4 achieved a Top1 diagnostic concordance in 108 (70.8%), Top3 concordance in 137 (87.7%), partial concordance in four (2.6%), and was discordant in 15 (9.7%) cases. The quality of ChatGPT-4's image descriptions significantly surpassed teledermatologists in all five parameters. ChatGPT-4's descriptions were accurate in 130 (84.4%), partially accurate in 22 (14.3%), and inaccurate in two (1.3%) cases. CONCLUSIONS: The preliminary findings of this study indicate that ChatGPT-4 demonstrates potential in generating accurate image descriptions and differential diagnoses. These results highlight the promise of integrating artificial intelligence into asynchronous teledermatology workflows.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.002
metaresearch head score (Gemma)0.002
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesMeta-epidemiology (narrow)
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Observational · Consensus signal: Observational
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.283
Threshold uncertainty score1.000

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0020.002
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0010.000
Bibliometrics0.0010.001
Science and technology studies0.0010.000
Scholarly communication0.0000.000
Open science0.0000.000
Research integrity0.0000.001
Insufficient payload (model declined to judge)0.0000.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.166
GPT teacher head0.411
Teacher spread0.245 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one teacher head, not a consensus.

Study designObservational
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations3
Published2025
Admission routes1
Has abstractyes

Explore more

Same venueJDDG Journal der Deutschen Dermatologischen GesellschaftSame topicArtificial Intelligence in Healthcare and EducationFrench-language works237,207