Improving Trauma Triage Accuracy with Large Language Models: A Comparison to Human Expert Decisions

Key Points

• To assess the accuracy of large language models in trauma triage compared to human experts, particularly in pediatric cases.
• Conducted a retrospective review of pediatric trauma cases.
• Compared triage accuracy between large language models and trauma staff decisions.
• Evaluated the potential for clinical outcome validation and user acceptance.
• Large language models achieved triage accuracy similar to that of trauma staff.
• Further validation is necessary for generalizability and user acceptance in real-world settings.

Abstract