Artificial Intelligence in Patch Testing: Comprehensive Review of Current Applications and Future Prospects in Dermatology
Bibliographic record
Abstract
Background: The integration of artificial intelligence (AI) into patch testing for allergic contact dermatitis (ACD) holds the potential to standardize diagnoses, reduce interobserver variability, and improve overall diagnostic accuracy. However, the challenges and limitations hindering clinical implementation have not been thoroughly explored. Objective: This narrative review aims to examine the current applications of AI in patch testing, identify challenges, and propose future directions for their use in dermatology. Methods: PubMed was searched in August 2024 to identify studies involving human participants undergoing patch testing with AI used in the study. Exclusion criteria were non-English and nonoriginal research. Data were synthesized to assess study design, performance, and potential for clinical application. Results: Out of 94 reviewed articles, 10 met the inclusion criteria. Most studies employed convolutional neural networks (CNN) for image analysis, with accuracy rates ranging from 90.1% to 99.5%. Other AI models, such as gradient boosting and random forest, were used for risk prediction and biomarker discovery. Key limitations included limited sample sizes, variability in image capture protocols, and lack of standardized reporting on skin types. Conclusions: AI has significant potential to enhance diagnostic accuracy, standardize patch test interpretation, and expand access to patch testing. However, standardized imaging protocols, larger and more diverse datasets, and improved regulatory frameworks are necessary to realize the full potential of AI in patch testing.
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 | 0.002 | 0.005 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.002 | 0.002 |
| Bibliometrics | 0.004 | 0.004 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.002 | 0.001 |
| 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".