Performance of Artificial Intelligence Imaging Models in Detecting Dermatological Manifestations in Higher Fitzpatrick Skin Color Classifications
Bibliographic record
Abstract
BACKGROUND: The performance of deep-learning image recognition models is below par when applied to images with Fitzpatrick classification skin types 4 and 5. OBJECTIVE: The objective of this research was to assess whether image recognition models perform differently when differentiating between dermatological diseases in individuals with darker skin color (Fitzpatrick skin types 4 and 5) than when differentiating between the same dermatological diseases in Caucasians (Fitzpatrick skin types 1, 2, and 3) when both models are trained on the same number of images. METHODS: Two image recognition models were trained, validated, and tested. The goal of each model was to differentiate between melanoma and basal cell carcinoma. Open-source images of melanoma and basal cell carcinoma were acquired from the Hellenic Dermatological Atlas, the Dermatology Atlas, the Interactive Dermatology Atlas, and DermNet NZ. RESULTS: The image recognition models trained and validated on images with light skin color had higher sensitivity, specificity, positive predictive value, negative predictive value, and F1 score than the image recognition models trained and validated on images of skin of color for differentiation between melanoma and basal cell carcinoma. CONCLUSIONS: A higher number of images of dermatological diseases in individuals with darker skin color than images of dermatological diseases in individuals with light skin color would need to be gathered for artificial intelligence models to perform equally well.
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 | 0.003 | 0.007 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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".