Related Experiment Video
Updated: Jan 15, 2026
A Large Animal Model for Acute Kidney Injury by Temporary Bilateral Renal Artery Occlusion
Published on: February 2, 2021
Predicting outcomes in pediatric patients with acute kidney injury: a retrospective single-center cohort study using
Feifei Shen1, Ying Xu2, Xusheng Jiang3
1Department of Pediatrics, Affiliated Hospital of Nantong University, Nantong, China.
Insights
Machine learning models accurately predict mortality in critically ill children with acute kidney injury (AKI). Elevated lactate levels are a key predictor, guiding early interventions for better outcomes.
Area of Science:
- Pediatric Critical Care Medicine
- Machine Learning in Healthcare
- Renal Medicine
Background:
- Acute kidney injury (AKI) is a significant concern in critically ill children, associated with high mortality rates.
- Accurate prediction of mortality is crucial for timely intervention and improved patient outcomes.
- Existing prediction models may not fully leverage advanced machine learning techniques for complex pediatric critical care data.
Purpose of the Study:
- To develop and evaluate machine learning (ML) models integrated with survival analysis for predicting 7-, 14-, and 28-day mortality in pediatric patients with AKI.
- To identify key predictors of mortality to facilitate risk stratification and guide early therapeutic strategies.
- To assess the temporal impact of predictors on mortality using time-to-event analyses.
Main Methods:
- Utilized the Pediatric Intensive Care (PIC) database, analyzing data from 3,624 children with AKI (2010-2018).
- Trained nine ML algorithms, including CatBoost, for mortality prediction; feature importance was determined using SHapley Additive exPlanations (SHAP).
- Conducted time-to-event analyses (Kaplan-Meier, restricted cubic splines) to examine predictor impact on 28-day mortality, stratified by age and AKI stage.
Main Results:
- CatBoost demonstrated superior performance with high Area Under the Curve (AUC) values: 0.871 (7-day), 0.871 (14-day), and 0.867 (28-day).
- Lactate emerged as the most significant predictor across all models.
- Time-to-event analysis showed a linear association between elevated lactate (>1.5 mmol/L) and 28-day mortality (p<0.001), particularly in infants and AKI stage 1 patients.
Conclusions:
- Machine learning, specifically CatBoost, combined with survival analysis, provides accurate prediction of mortality in critically ill children with AKI.
- Lactate is a critical marker for risk stratification and warrants targeted early interventions.
- Findings support precision medicine approaches, but multicenter validation is necessary for widespread clinical implementation.
Objective:
To develop and evaluate machine learning models combined with survival analysis for predicting 7-, 14-, and 28-day mortality in critically ill children with acute kidney injury (AKI), identifying key predictors to guide risk stratification and early intervention.
Methods:
Using the Pediatric Intensive Care (PIC) database, we analyzed data from 3,624 children with AKI admitted between 2010 and 2018. Nine machine learning algorithms, including CatBoost, were trained to predict mortality, with feature importance assessed via SHapley Additive exPlanations (SHAP). Time-to-event analyses, including Kaplan-Meier and restricted cubic spline methods, examined the temporal impact of predictors on 28-day mortality, stratified by age and AKI stage.
Results:
CatBoost achieved the highest area under the curve (AUC) values: 0.871 (95% CI: 0.824-0.918) for 7-day, 0.871 (95% CI: 0.829-0.913) for 14-day, and 0.867 (95% CI: 0.829-0.905) for 28-day mortality. Lactate was the top predictor across all models. Time-to-event analyses revealed a linear association between elevated lactate (cut-off: 1.5 mmol/L) and 28-day mortality (p-overall < 0.001), with stronger effects in infants (0-3 years) and AKI stage 1 patients (HR > 1).
Conclusions:
Machine learning, particularly CatBoost, combined with survival analysis, accurately predicts AKI-related mortality in critically ill children, with lactate as a pivotal marker. These findings support precision risk stratification and early lactate-targeted interventions, though multicenter validation is needed for clinical adoption.
Related Concept Videos
01:30Acute Kidney Injury IV: Diagnostic Studies and Prevention
01:22Acute Kidney Injury I: Introduction
01:29Acute Kidney Injury II: Pathophysiology
01:20Acute Kidney Injury V: Interprofessional Care
01:29Acute Kidney Injury III: Clinical Manifestations
01:28Drug Dosing in Renal Diseases: Estimation of Glomerular Filtration Rate Based on Serum Creatinine Concentration