EUVIMEDEUVIMED entdecken
Crossref

Diagnostic Accuracy of Large Language Models for Fictional Pediatric Neuroradiology Cases: A Comparative Study with Radiologists

Turay Cesur, Yasin Gunes, Eren Camur, Gulay Cinar, Goksel Tuzcu, Avni Keceli

Proceedings of the Bulgarian Academy of Sciences · 29.05.2026 · 79 (5)

DOI: 10.7546/crabs.2026.05.12 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the Bulgarian Academy of Sciences, online erschienen am 29.05.2026. Verfasst von Turay Cesur und 5 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The aim of this study was to evaluate diagnostic accuracy and differential-diagnosis quality of large language models (LLMs) for clinically realistic, text-only pediatric neuroradiology cases. This cross-sectional diagnostic accuracy study included 100 fictional pediatric neuroradiology cases composed of a brief clinical presentation and a structured text-only CT/MRI report curated by an expert pediatric neuroradiologist and a pediatrist. For each case, a reference primary diagnosis and two acceptable alternatives were pre-specified. Seven LLM variants (ChatGPT-5.2 Instant/Auto/Thinking; Gemini 3 Pro/Thinking; Claude 4.5 Opus/Opus Thinking) and three radiologists (two general radiologists; one pediatric radiologist) each provided one primary diagnosis and two differentials. Primary outcome was top-1 accuracy (exact match or acceptable alternative). Secondary outcomes were Differential Diagnosis Score (DDxScore, 1--5) and response time. Paired accuracy differences were assessed with Cochran's Q and post-hoc McNemar tests; DDxScore and response time were compared using Friedman tests with post-hoc Wilcoxon signed-rank tests and multiplicity correction. Top-1 accuracy ranged from 44–54% among radiologists (pediatric radiologist 54%) and 48–80% among LLMs (ChatGPT-5.2 Thinking/Auto and Gemini 3 Thinking 80%; Claude 4.5 Opus Thinking 76%). Overall accuracy differed across raters (Cochran's Q = 107.86, df = 9, p < 0.001). Median DDxScores were 4.0 (IQR 2.0–4.0) for the pediatric radiologist, 3.0 (2.0–4.0) for general radiologists, and up to 5.0 (4.0–5.0) for leading thinking-mode LLMs (p < 0.001). All LLMs were faster than radiologists (p < 0.001). In text-only pediatric neuroradiology cases, several contemporary LLMs matched or exceeded radiologists in top-1 accuracy and produced high-quality differentials with substantially shorter response times. These findings support further evaluation for education and audited decision-support workflows.

Zitieren

APA

Cesur, T., Gunes, Y., Camur, E., Cinar, G., Tuzcu, G., & Keceli, A. (2026). Diagnostic Accuracy of Large Language Models for Fictional Pediatric Neuroradiology Cases: A Comparative Study with Radiologists. Proceedings of the Bulgarian Academy of Sciences, 79(5). https://doi.org/10.7546/crabs.2026.05.12

Vancouver

Cesur T, Gunes Y, Camur E, Cinar G, Tuzcu G, Keceli A. Diagnostic Accuracy of Large Language Models for Fictional Pediatric Neuroradiology Cases: A Comparative Study with Radiologists. Proceedings of the Bulgarian Academy of Sciences. 2026;79(5). doi:10.7546/crabs.2026.05.12

BibTeX
@article{Cesur2026,
  author = {Cesur, Turay and Gunes, Yasin and Camur, Eren and Cinar, Gulay and Tuzcu, Goksel and Keceli, Avni},
  title = {Diagnostic Accuracy of Large Language Models for Fictional Pediatric Neuroradiology Cases: A Comparative Study with Radiologists},
  journal = {Proceedings of the Bulgarian Academy of Sciences},
  year = {2026},
  volume = {79},
  number = {5},
  doi = {10.7546/crabs.2026.05.12}
}
Kostenlos starten →