Scenario-Based Prehospital Triage of Carbon Monoxide Poisoning: Comparing Five Large Language Models With Emergency Medical Services Personnel in a Prospective Study

Vildan Özer; Ozlem Bulbul; Efnan Bayrak Erbolukbas; Serdar Karakullukçu; Aynur Şahin

doi:10.24938/kutfd.1922944

August 12, 2026Kırıkkale Üniversitesi Tıp Fakültesi Dergisi

Scenario-Based Prehospital Triage of Carbon Monoxide Poisoning: Comparing Five Large Language Models With Emergency Medical Services Personnel in a Prospective Study

View Full Paper

Ask AI

Bookmark

Share

Key Points

To compare the decision-making accuracy of emergency medical services personnel and five large language models in determining the need for prehospital hyperbaric oxygen therapy in carbon monoxide poisoning.
Prospective, simulation-based diagnostic accuracy study evaluating 128 standardized scenario-based clinical cases (64 requiring hyperbaric oxygen therapy, 64 not requiring it).
Responses from 60 emergency medical services personnel and five large language models (GPT-4o, GPT-4.5, GPT-o3, Gemini 2.5 Pro, and DeepSeek-R1) were benchmarked against a specialist consensus gold standard.
Emergency medical services personnel achieved an overall diagnostic accuracy of 65.1%, whereas GPT-o3 achieved the highest accuracy at 96.9%.
All five large language models demonstrated 100% sensitivity for identifying the need for hyperbaric oxygen therapy, with specificity ranging from 26.6% (GPT-4.5) to 93.8% (GPT-o3).

No abstract yet

We only have this paper's title and publication details.

Read it at the source

Ask about this paper

Answers are grounded in this paper and cite their sources.

Implication

Prospective simulation study reveals superior triage accuracy of LLMs over paramedics in carbon monoxide poisoning, highlighting their potential as clinical decision-support tools.

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Özer et al. (2026) studied this question.

synapsesocial.com/papers/6a7ec735b70b84ec8b9134e3 https://doi.org/10.24938/kutfd.1922944

View Full Paper

Ask AI

Bookmark

Share

Scenario-Based Prehospital Triage of Carbon Monoxide Poisoning: Comparing Five Large Language Models With Emergency Medical Services Personnel in a Prospective Study

Key Points

No abstract yet

Ask about this paper

Implication

Discussion

Cite This Study

Scenario-Based Prehospital Triage of Carbon Monoxide Poisoning: Comparing Five Large Language Models With Emergency Medical Services Personnel in a Prospective Study

Key Points

No abstract yet

Ask about this paper

Implication

Discussion

Cite This Study

Keep reading

Keep reading