Efficacy of large language models in detecting postoperative delirium from unstructured clinical notes: A retrospective cohort study
Da-In Eun, Hyung-Chul Lee, Gang Heo, Jae Hoon Jeong, Jipyeong Lee, Hyeonhoon Lee, Sang-Youn Park, Karam Nam, Ho-Jin Lee, Jae-Woo Ju · 2025
Read Efficacy of large language models in detecting postoperative delirium from unstructured clinical notes: A retrospective cohort study by Da-In Eun, Hyung-Chul Lee, Gang Heo, Jae Hoon Jeong, Jipyeong Lee, Hyeonhoon Lee, Sang-Youn Park, Karam Nam, Ho-Jin Lee, Jae-Woo Ju (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-12816606.txt.
Description
Da-In Eun and nine co-authors test whether large language models can spot delirium after surgery by reading doctors' notes, in a 2025 study. Delirium is common, harmful, and often missed or caught late.
Two models, Llama-3-70B and GPT-4o, were compared with physicians at flagging serious cases, meaning those needing drugs or a specialist's diagnosis. The models caught more true cases than physicians, about 87 to 90 percent against 72 percent, but raised more false alarms. The two models agreed with each other almost perfectly, while physicians agreed only fairly. Both models flagged cases about a day earlier than doctors. The authors suggest models as a screening aid under doctor supervision.
Catalog record
| CREATOR | Da-In Eun, Hyung-Chul Lee, Gang Heo, Jae Hoon Jeong, Jipyeong Lee, Hyeonhoon Lee, Sang-Youn Park, Karam Nam, Ho-Jin Lee, Jae-Woo Ju |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 28 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |