LLMs Outperform Clinicians at Crafting Clinical Context for Imaging Orders, Study Finds

LLM-generated clinical histories for imaging orders were more comprehensive and factual than those written by referring clinicians. Claude 3.5 Sonnet was rated most useful for protocoling and interpretation, per a Radiology study of >28,000 records.

• Twenty radiologists compared indications from large language models and clinicians; LLM-produced histories scored higher on comprehensiveness and factual accuracy.
• Comprehensiveness was the single most important factor influencing overall usefulness of an imaging indication.

Automated summary

RadPigeon summaries are original and for information only. They are not clinical advice.