2026-07-10 バッファロー大学（UB）
The chaos of an accident scene, along with noisy radio connections and time pressure are some of the factors that can create fertile ground for miscommunication.
＜関連情報＞
- https://www.buffalo.edu/news/releases/2026/07/Trauma-triage-can-LLM-help-UB-Surgery.html
- https://www.ovid.com/jnls/journalacs/abstract/10.1097/xcs.0000000000001895~improving-trauma-triage-accuracy-with-large-language-models?redirectionsource=fulltextview
大規模言語モデルを用いた外傷トリアージ精度の向上：人間の専門家の判断との比較 Improving Trauma Triage Accuracy with Large Language Models: A Comparison to Human Expert Decisions
Ascharya K Balaji,Brendan T Fox,Philip Seger,Akhil Gorugantu,Andrew Nordin,Tiffany Fabiano,Gene Yang,Sharifa Himidan,Steven D Schwaitzberg,Peter CW Kim
Journal of the American College of Surgeons Accepted: 13 January 2026
DOI: 10.1097/XCS.0000000000001895
Abstract
Background:
Accurate prehospital trauma triage and communication determine morbidity, mortality, and system efficiency. Advancements in large language models (LLMs) offer an opportunity to improve triage and yet to be implemented in prehospital trauma triage.
Study Design:
This retrospective cohort study evaluates LLM performance in trauma triage and accuracy of prehospital tele-communication. Of 410 pediatric activations at a Level I center (January 2023 to May 2025, IRB no. 00009569), 133 activations with emergency medical service recordings, human-generated trauma pages, and injury severity scores were analyzed. Audio was transcribed with OpenAI Whisper. Structured “Essential Transcripts” were generated with named entity recognition. Entity ablation tested redaction of triage parameters on accuracy. In a prospective arm, trauma surgeons reviewed emergency medical service transcripts and triaged activations pre- and post-LLM exposure. Cribari criteria defined over and undertriage. McNemar’s test and 95% CIs assessed paired differences in accuracy.
Results:
The primary endpoint: LLM undertriage demonstrated modest improvement; 4.8% (3 to 8.2) vs 5.1% (3.1 to 9.3, p = 0.73, Bonferroni p = 1). For secondary endpoints, LLM triage outperformed human clinicians: 83.5% accuracy (80.6 to 90.6) vs 78.9% (73.9 to 82.6, p < 0.01, Bonferroni p < 0.01), with overtriage 58.6% (51.4 to 73.7) vs 71.8% (p < 0.05, Bonferroni p < 0.09). “Essential Transcripts” reduced transcript length by 80.8% (p < 0.001) while preserving accuracy (81.9%, 76.6 to 87.5, p < 0.001, Bonferroni p < 0.05). Entity ablation had marginal effect on triage. In prospective evaluation, human triage accuracy improved after LLM exposure (73.7% [69.8 to 77.2]) to 75.8% ([71.9 to 79.2], p = 0.04, Bonferroni p = 0.12), significantly improving the odds of a correct triage decision (odds ratio 2.57, 95% CI 1.39 to 6.83, p < 0.01).
Conclusions:
LLMs achieve triage accuracy comparable to trauma staff in retrospective review of pediatric trauma. Further validation is needed to assess clinical outcomes, generalizability, and user acceptance before widespread deployment.
