Distinguishing Extraction Failures and Clinical Discordance with LLM Uncertainty on Radiology Reports

Jerjes Aguirre-Chavez, Shant Malkasian, Roshun Sankaran +2

Large language models (LLMs) can structure clinical datasets, but extraction error must be distinguished from discordance in the underlying data. From a real-world cohort of 1{,}04…