Related Experiment Video
Updated: May 5, 2026
A Large Animal Model for Acute Kidney Injury by Temporary Bilateral Renal Artery Occlusion
Published on: February 2, 2021
Artificial intelligence for predicting paediatric acute kidney injury: a systematic review and meta-analysis
Rupesh Raina1,2,3, Parth Shirode1,2, Raghav Shah3
1Department of Pediatric Nephrology, Akron Children's Hospital, Akron, OH, USA.
Insights
Artificial intelligence (AI) and machine learning (ML) show promise for predicting acute kidney injury (AKI) in children. Gradient boosting models achieved the highest predictive accuracy, though further validation is needed.
Area of Science:
- Pediatric Nephrology
- Medical Informatics
- Artificial Intelligence in Healthcare
Background:
- Acute kidney injury (AKI) is a significant complication in hospitalized children, leading to increased morbidity and mortality.
- Early detection and risk stratification of pediatric AKI are crucial for improving patient outcomes.
- Artificial intelligence (AI) and machine learning (ML) offer potential tools for enhancing AKI prediction in pediatric populations.
Purpose of the Study:
- To systematically review and evaluate the performance of AI/ML models for predicting pediatric AKI.
- To assess the effectiveness of various AI/ML algorithms in identifying children at risk of AKI.
- To identify the most promising AI/ML approaches for pediatric AKI prediction in clinical settings.
Main Methods:
- A systematic literature search was conducted across PubMed, Embase, and Web of Science.
- Studies utilizing AI/ML models for pediatric AKI prediction were included if they reported key performance metrics.
- Performance metrics included area under the curve (AUC), sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), accuracy, and F1 score.
Main Results:
- Eleven studies involving 14 AI/ML models and 33,949 pediatric patients were included.
- Gradient boosting models achieved the highest pooled AUC (0.873), indicating strong predictive performance.
- Random forest models showed high median sensitivity, specificity, PPV, NPV, and accuracy, but results were not pooled due to data heterogeneity.
Conclusions:
- AI/ML models, particularly gradient boosting and random forest, demonstrate potential for predicting pediatric AKI.
- Limitations such as small sample sizes, data heterogeneity, and inconsistent diagnostic criteria hinder the generalizability of current models.
- Further research with larger, validated cohorts is necessary to refine and implement these AI/ML tools for clinical use.
Background:
Acute kidney injury (AKI) in hospitalised children is a major complication associated with significant morbidity and mortality. The integration of artificial intelligence (AI)/machine learning (ML) models may enable early detection and risk stratification. This systematic review evaluates the performance of AI/ML models for predicting paediatric AKI across clinical settings.
Methods:
We systematically searched PubMed, Embase, and Web of Science for studies applying AI/ML models to predict AKI in paediatric populations. Studies reporting performance metrics such as the area under the curve (AUC), sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), accuracy, and F1 score were included.
Results:
Among 470 records identified, 11 studies met the inclusion criteria, with 14 AI/ML models used. The overall sample size included 33 949 paediatric patients with an AKI proportion of 12.5%. Meta-analyses of the AUC were conducted on neural network, gradient boosting, and logistic regression. Gradient boosting had the highest pooled AUC of 0.873 (95% confidence interval 0.836-0.909). Random forest demonstrated the highest median sensitivity (0.821), specificity (0.942), PPV (0.860), NPV (0.935), and accuracy (0.821); however, these metrics could not be pooled due to inconsistent reporting and limited validation.
Conclusion:
Gradient boosting, random forest, and logistic regression demonstrated reasonable predictive performance for paediatric AKI prediction within specific clinical contexts. However, small sample size, heterogeneity, lack of testing/validation cohorts, insufficient data, and inconsistent patient populations and AKI diagnostic criteria restrict generalisability.
Related Concept Videos
01:30Acute Kidney Injury IV: Diagnostic Studies and Prevention
01:22Acute Kidney Injury I: Introduction
01:20Acute Kidney Injury V: Interprofessional Care
01:29Acute Kidney Injury II: Pathophysiology
01:29Acute Kidney Injury III: Clinical Manifestations
01:22Acute Kidney Injury VI: Nursing Management