Related Experiment Video
Updated: May 10, 2025
A Large Animal Model for Acute Kidney Injury by Temporary Bilateral Renal Artery Occlusion
Published on: February 2, 2021
Development of a neural network model for early detection of creatinine change in critically Ill children
Celeste G Dixon1, Eduardo A Trujillo Rivera1, Anita K Patel1
1Department of Pediatrics, Division of Critical Care Medicine, Children's National Hospital, George Washington University School of Medicine and Health Sciences, Washington, DC, United States.
Insights
Machine learning predicts 24-hour creatinine change in critically ill children, identifying renal dysfunction risk early. This aids timely intervention before clinical detection, improving outcomes for pediatric intensive care unit patients.
Area of Science:
- Pediatric Nephrology
- Critical Care Medicine
- Biomedical Informatics
Background:
- Renal dysfunction is a significant concern in critically ill children, increasing morbidity and mortality.
- Current diagnosis relies on creatinine, a marker with delayed response to renal injury.
- Early prediction of renal dysfunction is crucial for timely intervention.
Purpose of the Study:
- To develop and validate a machine learning model to predict 24-hour creatinine change in critically ill children.
- To identify children at risk of significant renal dysfunction before it is clinically apparent.
Main Methods:
- Retrospective cohort study of 39,932 pediatric intensive care unit encounters.
- A neural network model was trained using demographics, vital signs, lab tests, and medications.
- The model predicted <50% or ≥50% creatinine change within 24 hours.
Main Results:
- The model achieved 68.1% overall accuracy in predicting creatinine change.
- Prediction accuracy improved significantly with higher admission creatinine levels (up to 96.3%).
- The model demonstrated a negative predictive value of 97.2% for detecting significant creatinine change.
Conclusions:
- Machine learning models can predict 24-hour creatinine change using routine clinical data in critically ill children.
- This predictive capability offers a window for early risk identification of renal dysfunction.
- Clinical utility is influenced by the reliance on creatinine as a diagnostic marker.
Introduction:
Renal dysfunction is common in critically ill children and increases morbidity and mortality risk. Diagnosis and management of renal dysfunction relies on creatinine, a delayed marker of renal injury. We aimed to develop and validate a machine learning model using routinely collected clinical data to predict 24-hour creatinine change in critically ill children before change is observed clinically.
Methods:
Retrospective cohort study of 39,932 pediatric intensive care unit encounters in a national multicenter database from 2007 to 2022. A neural network was trained to predict <50% or ≥50% creatinine change in the next 24 h. Admission demographics, routinely measured vital signs, laboratory tests, and medication use variables were used as predictors for the model. Data set was randomly split at the encounter level into model development (80%) and test (20%) sets. Performance and clinical relevance was assessed in the test set by accuracy of prediction classification and confusion matrix metrics.
Results:
The cohort had a male predominance (53.8%), median age of 8.0 years (IQR 1.9-14.6), 21.0% incidence of acute kidney injury, and 2.3% mortality. The overall accuracy of the model for predicting change of <50% or ≥50% was 68.1% (95% CI 67.6%-68.7%). The accuracy of classification improved substantially with higher creatinine values from 29.9% (CI 28.9%-31.0%) in pairs with an admission creatinine <0.3 mg/dl to 90.0-96.3% in pairs with an admission creatinine of ≥0.6 mg/dl. The model had a negative predictive value of 97.2% and a positive predictive value of 7.1%. The number needed to evaluate to detect one true change ≥50% was 14.
Discussion:
24-hour creatinine change consistent with acute kidney injury can be predicted using routine clinical data in a machine learning model, indicating risk of significant renal dysfunction before it is measured clinically. Positive predictive performance is limited by clinical reliance on creatinine.
