Fig. 3From: Automated extraction of functional biomarkers of verbal and ambulatory ability from multi-institutional clinical notes using large language modelsProportion and correctness of non-unknown GPT note-level BGR predictionsBack to article page
