Use of a Large Language Model to Identify and Classify Injuries With Free-Text Emergency Department Data

This cross-sectional study assesses the accuracy, sensitivity, and specificity of a large language model used to process unstructured, non-English emergency department (ED) data in medical records.