Generalization of finetuned transformer language models to new clinical contexts.
Kevin XieSamuel W TermanRyan S GallagherChloe E HillKathryn A DavisBrian LittDan RothColin A EllisPublished in: JAMIA open (2023)
Model generalization performance decreased on notes from nonspecialists; out-of-institution generalization on epilepsy specialist notes required small changes to preprocessing but was especially good for seizure frequency text and date of last seizure text, opening opportunities for multicenter collaborations using these outcomes.