Enhancing Intelligent Triage with Large Language Models: A Comprehensive Evaluation and Optimization Study

Abstract

This study explores the application of LLMs in intelligent triage tasks with the aim of enhancing patient triage efficiency and accuracy, as well as optimizing healthcare resource allocation. A variety of LLMs were assessed across different triage scenarios. Through prompt engineering, parameter tuning, and workflow design, optimal configurations were identified, significantly boosting triage accuracy. The research highlights the significance of customizing LLMs for specific healthcare requirements and showcases their potential to deliver high-precision triage assistance in real-world settings. The findings indicate that well-designed LLMs can greatly improve triage accuracy and adapt to diverse scenarios, ultimately enhancing patient care and resource allocation. Our code has been released on GitHub*.

Ähnliche Arbeiten