Large language models facilitate the generation of electronic health record phenotyping algorithms.

Chao YanHenry H OngMonika E GrabowskaMatthew S KrantzWu-Chen SuAlyson L DicksonJosh F PetersonQiPing FengDan M RodenC Michael SteinVern Eric KerchbergerBradley A MalinWei-Qi Wei

Published in: Journal of the American Medical Informatics Association : JAMIA (2024)

GPT versions 3.5 and 4 are capable of drafting phenotyping algorithms by identifying relevant clinical criteria aligned with a CDM. However, expertise in informatics and clinical experience is still required to assess and further refine generated algorithms.

Keyphrases

electronic health record
machine learning
deep learning
high throughput
clinical decision support
big data
artificial intelligence
adverse drug
autism spectrum disorder