Perioperative medicine & outcomes
A pre-trained language model approach for triaging surgical patients for preoperative anesthesia clinics
Authors' conclusion
Automating integrated analysis using pre-trained language models to aid in preoperative triaging could enhance accuracy and efficiency at scale, reducing manual review and provider burden.
Study context
- Population
- Surgical patients evaluated at a single-centre preoperative anesthesia care clinic; classification of nursing preoperative phone call vs in-person clinician evaluation46,922 clinical documents
- Participants
- 1,761 patients
- Comparison
- Four pre-trained language models with three data inputs
- Outcome
- AUC range across the four language models
Methodological detailssingle-centre, machine learning (pre-trained language models), 75%/25% training/test split
- Single-centre
- Machine learning (pre-trained language models)
- 75%/25% training/test split
Performance
As reported in the abstract
| Test or model | AUC |
|---|---|
| D + N | 0.70 - 0.74 |
| N | 0.58 - 0.73 |
| D | 0.57 - 0.62 |
AUC
D + N0.70 - 0.74
N0.58 - 0.73
D0.57 - 0.62
Original publication
A pre-trained language model approach for triaging surgical patients for preoperative anesthesia clinics. J Clin Monit Comput. Published 24 December 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →