Pediatric anesthesia & critical care
Assessing the Utility of a Machine-Learning Model to Assist With the Assignment of the American Society of Anesthesiology Physical Status Classification in Pediatric Patients
Authors' conclusion
ML derivation of predicted pediatric ASA-PS scores was successful, with a strong agreement between predicted and clinician-entered ASA-PS scores. Presentation of predicted ASA-PS scores was associated with revision in final scoring for 1-in-10 pediatric patients.
Study context
- Population
- Pediatric patients undergoing elective surgical procedures
- Participants
- 146,784 retrospective encounters
- Exposure
- Machine-learning (ML)-predicted vs clinician-assigned ASA-PS scores
- Outcome
- Distribution of ASA-PS scores (retrospective)
Methodological detailsmachine-learning (XGBoost), SHAP
- Machine-learning (XGBoost)
- SHAP
Main finding
As reported in the abstract
ASA-PS I
II
III
IV
V
Observational data: associations, not effects.
Further findings
Final vs initial clinician ASA-PS after viewing prediction (prospective)
- Final matched initial ASA-PS
- 90.7%
- Revised after viewing predicted ASA-PS
- 9.3%
- Final differed from initial when initial and ML scores discrepant
- 19.5%
Prevalence of multiple chronic conditions by ASA-PS
- ASA-PS I
- 34.9%
- ASA-PS II
- 73.2%
- ASA-PS III
- 92.3%
- ASA-PS IV
- 94.4%
Original publication
Assessing the Utility of a Machine-Learning Model to Assist With the Assignment of the American Society of Anesthesiology Physical Status Classification in Pediatric Patients. Anesth Analg. Published 13 December 2023. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →