Emerging Uses of Artificial Intelligence in Chronic Dermatologic Disease: A Scoping Review
Bibliographic record
Abstract
BACKGROUND: Recent years have seen a surge in the use of artificial intelligence (AI) in healthcare, including dermatology. This scoping review aimed to assess the emerging applications of AI use in the context of chronic, non-neoplastic dermatologic diseases. METHODS: MEDLINE, Embase, PubMed and SCOPUS were searched on August 11, 2023 using variations of the search concepts "dermatology," "artificial intelligence," and 12 common chronic dermatologic conditions. Article screening and data extraction were completed, and each study was categorized into themes and conditions. RESULTS: A total of 224 unique studies were included. The most prevalent conditions that were studied in the context of AI included psoriasis (n = 67), atopic dermatitis/eczema (n = 41) and acne (n = 36). The majority of AI applications involved clinical evaluation (n = 176), images (analysis, generation or segmentation) (n = 163) and data analysis (n = 46). Clinical evaluation was further divided into 2 subthemes: diagnosis (n = 104) and disease assessment (n = 67). Diagnostic and analytic applications of AI are limited by the training datasets available (quantity of training data, image quality) and insufficient diagnostic information provided (eg, the patient's reported history of their lesion, disease/symptom onset and risk factors). CONCLUSIONS: Common applications of AI are predominantly as an automated diagnostic tool for evaluating disease severity/characteristics, while niche and novel applications were explored further. However, recognizing the limitations of technology is critical prior to the widespread application of AI in dermatological practice. The insights from the current study can inform clinical adoption of AI in dermatology, and highlight research gaps to guide future academic initiatives.
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 | 0.012 | 0.047 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.003 | 0.004 |
| Bibliometrics | 0.023 | 0.020 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.005 | 0.004 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.003 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 0.000 |
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.
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".