Paediatric anaesthesia & critical care
Machine Learning Modeling for Predicting Mortality in Pediatric Patients Undergoing Elective Noncardiac Surgery: Comparison to a Regression Model
Authors' conclusion
ML can be leveraged to develop clinical prediction tools with excellent predictive performance for rare but critical outcomes like postoperative mortality. However, the ML-based models in the current study performed similarly to the regression-based PRAm score
Study context
- Population
- Patients <18 years undergoing multispecialty noncardiac surgery, NSQIP-Pediatric 2012–2022 excluding 20203,522 (0.34%) died
- Participants
- 1,023,639 patients
- Exposure
- XGBoost and random forest machine learning (ML) models vs regression-based Pediatric Risk Assessment (PRAm) score
- Outcome
- 30-day mortality
- Follow-up
- 30 days
Methodological detailsprediction model development and internal validation, NSQIP-Pediatric database
- Prediction model development and internal validation
- NSQIP-Pediatric database
Main finding
As reported in the abstract
Prediction of 30-day mortality (validation set)
| Group | AUC-ROC | AUC-PR | Accuracy | Precision | Brier score |
|---|---|---|---|---|---|
| XGBoost | 0.956 | 0.179 | 99.4% | 0.247 | 0.003 |
| PRAm | 0.958 |
AUC-ROC
XGBoost0.956
PRAm0.958
AUC-PR
XGBoost0.179
PRAm— not stated in the abstract
Accuracy
XGBoost99.4%
PRAm— not stated in the abstract
Precision
XGBoost0.247
PRAm— not stated in the abstract
Brier score
XGBoost0.003
PRAm— not stated in the abstract
Observational data: associations, not effects.
Original publication
Machine Learning Modeling for Predicting Mortality in Pediatric Patients Undergoing Elective Noncardiac Surgery: Comparison to a Regression Model. Anesth Analg. Published 25 August 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →