Pediatric anesthesia & critical care
A machine‐learning approach for decision support and risk stratification of pediatric perioperative patients based on the APRICOT dataset
Authors' conclusion
This work demonstrates that prediction of patients at low risk of critical PAEs can be made on an individual, rather than population-based, level by using machine learning.
Study context
- Population
- Children, American Society of Anesthesiologists Physical Status (ASA-PS) I to III, first procedure, APRICOT cohort (261 European institutions, 2014–2015)
- Participants
- 30,325 records
- Primary endpoint
- Classification of children at low risk for severe perioperative critical events (respiratory, cardiac, allergic, neurological), at surgical booking and on the procedure day
Methodological detailsmachine learning, secondary analysis of a prospective observational multicentre cohort, stratified 70:30 train:test split
- Machine learning
- Secondary analysis of a prospective observational multicentre cohort
- Stratified 70:30 train:test split
Primary result
As reported in the abstract
Model performance
Accuracy
>0.9
AUC
0.6-0.7
NPV
>95%
Further findings
Adverse event rate in dataset
- Adverse event rate
- 4.43%
Original publication
A machine‐learning approach for decision support and risk stratification of pediatric perioperative patients based on the APRICOT dataset. Paediatr Anaesth. Published 21 May 2023. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →