AI in Patient Care: Evaluating Large Language Model Performance Against Evidence-Based Guidelines for Pulmonary Embolism

Key Points

• Evaluate the performance of large language models (LLMs) in managing pulmonary embolism (PE) compared to evidence-based guidelines.
• Assessment of LLMs in relation to PE management guidelines
• Comparison of LLM performance across different domains of care
• Analysis of guideline compliance in AI-generated responses
• No single LLM consistently excelled across all evaluation domains
• LLMs demonstrated potential in supporting PE management
• Need for further development to improve clinical integration and adherence to guidelines