Evaluation of a Large Language Model to Identify Confidential Content in Adolescent Encounter Notes

Abstract

This cross-sectional study assesses the ability of a language learning model to classify whether a progress note contains confidential information and to identify the specific confidential content in the note.