Implementation and prospective performance evaluation of an intraoperative duration prediction model using high throughput real-time data
Authors' conclusion
Prospective evaluation showed that model performance was preserved over a 10-month testing period.
Study context
- Population
- Surgical procedures with real-time anaesthesia record data
- Participants
- 62,142 procedures
- Exposure
- Machine learning (ML) model vs bias-corrected scheduled duration
- Outcome
- Prediction of intraoperative duration, continuous ranked probability score
- Follow-up
- 10 months
Methodological detailssilent deployment, real-time data
- Silent deployment
- Real-time data
Main finding
As reported in the abstract
Continuous ranked probability score
Observational data: associations, not effects.
Further findings
Performance drift over testing period · 10 months
- Linear regression did not demonstrate performance drift
Original publication
Implementation and prospective performance evaluation of an intraoperative duration prediction model using high throughput real-time data. BJA Open. Published 7 May 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →