ArticlePediatric dermatology
Digitizing Diagnoses: Distinguishing Infantile Hemangiomas From Other Vascular Anomalies.
Article in Pediatric dermatology. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Cited by 1 paper.
What it found
Each row is one number read from the abstract, on the scale the paper reported it, with its interval. Left of the dashed line favours the treatment, right favours the comparator. Under each row is the sentence it came from. New to these charts? A ten-minute tutorial.
The abstract states no effect estimate the extractor could read, or names no intervention and outcome on the map, so this paper lights no cell and moves no belief. It is still indexed, cited and linked below.
The trial behind it
Trials whose registry record cites this paper, or whose number appears in the abstract. A trial that started after this paper was published is citing it as background, not reporting it.
Neither the registry nor the abstract names a trial number. If this is a trial report, that itself is worth knowing.
Who cites it
1 citing paper in PubMed.
- Digitizing Diagnoses: Distinguishing Infantile Hemangiomas From Other Vascular Anomalies.Pediatric dermatologyArticle
Corrections and comments
PubMed lists nothing against this paper. Absence here is not a guarantee, only a check that was made.
Authors and funding
7 authors.
Funding
No grant is acknowledged in the PubMed record.
Abstract
BACKGROUND/
objectivesGenerative artificial intelligence (AI) models have become increasingly accessible and advanced with multimodal input. Infantile hemangiomas (IHs) are the most common pediatric vascular tumor, but pediatricians may have difficulty distinguishing them from similar-appearing lesions. We assess the capability of a public AI model, ChatGPT 4.0 (GPT) to distinguish between IHs and other vascular anomalies (VAs), evaluating its potential role as a clinical tool for pediatric clinicians.
methodsThis retrospective study assessed 50 IH and 50 non-IH VA images using a GPT zero-shot approach with the binary task of diagnosing an IH (or not). The same images were provided to four general pediatricians; comparison was performed between pediatricians and GPT.
resultsGPT achieves 75% accuracy in IH identification with an F1 score of 0.742. ROC curve generation yields an AUC of 0.80. Hundred-image analysis demonstrates that actual diagnosis affects GPT accuracy (p = 0.015). 50-IH-image analysis reveals configuration (p = 0.027), skin phototype (p = 0.019), and anatomical location (p = 0.047) as factors that may affect GPT accuracy. Comparing GPT to pediatricians reveals comparable results (p = 0.345).
conclusionThis off-the-shelf publicly available GPT (75%) was less accurate than a previously published well-trained AI that achieved higher accuracy (~92%). GPT's F1 score of 0.742 indicates moderate balance between precision and sensitivity, and an AUC calculation of 0.80 indicates a potential role for GPT as a clinical assistant in primary care settings. An untuned commercial GPT does not yet outperform pediatricians despite studies of well-trained AI. This study provides variables impacting GPT accuracy, serving as a foundation for future research.
Indexed as
Identifiers
What OpenQuestion holds
Registered trials
Read under generation 80e0d062 · epoch 390. Bibliography from PubMed, PubMed Central and OpenAlex; grants from NIH RePORTER; trial links from ClinicalTrials.gov; estimates, votes and beliefs from the OpenQuestion graph.