Diagnosing Allergic Contact Dermatitis Using Deep Learning: Single-Arm, Pragmatic Clinical Trial with an Observer Performance Study to Compare Artificial Intelligence Performance with Human Reader Performance
Bibliographic record
Abstract
Abstract: Background: Allergic contact dermatitis is a common, pruritic, debilitating skin disease, affecting at least 20% of the population. Objective: To prospectively validate a computer vision algorithm across all Fitzpatrick skin types. Methods: Each participant was exposed to 10 allergens. The reference criterion was obtained 5 days after initial patch placement by a board-certified dermatologist. The algorithm processed photographs of the test site obtained on Day 5. Human performance in reading the photographs was also evaluated. Results: A total of 206 evaluable participants [mean age 39 years, 66% (136/206) female, and 47% with Fitzpatrick skin types IV–VI] completed testing. Forty-two percent (87/206) of participants experienced 1 or more allergic reaction resulting in a total of 132 allergic reactions. The model provided high discrimination (AUROC 0.86, 95% CI: 0.82–0.90) and specificity (93%, 95% CI: 92%–94%) but with lower sensitivity (58%, 95% CI: 49%–67%). Human performance interpreting the photographs ranged from providing similar performance to the algorithm to providing superior performance when combined across readers. There were no serious adverse events. Conclusions: The combination of a smartphone capture of patch testing sites with deep learning yielded high discrimination across a diverse sample.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.007 | 0.007 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.002 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
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.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
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".