<ul><li>Clinical Decision Rules (CDRs) are tools used in clinical decision-making to make consistent and accurate diagnoses by combining signs, symptoms, and clinical variables into decision trees.</li><li>CDR-Agent is a novel system based on Large Language Models (LLMs) designed to autonomously identify and apply appropriate CDRs based on unstructured clinical notes in emergency departments.</li><li>CDR-Agent achieved significant accuracy gains in CDR selection (56.3% in synthetic dataset and 8.7% in CDR-Bench) compared to standalone LLM baseline.</li><li>The system not only efficiently selects relevant CDRs but also makes effective imaging decisions, reducing unnecessary interventions while successfully identifying positively diagnosed cases.</li></ul>

CDR-Agent: Intelligent Selection and Execution of Clinical Decision Rules Using Large Language Model Agents

Discover more