Related Experiment Video
Updated: Jun 5, 2025
A Large Animal Model for Acute Kidney Injury by Temporary Bilateral Renal Artery Occlusion
Published on: February 2, 2021
Acute kidney disease in hospitalized pediatric patients: risk prediction based on an artificial intelligence approach
Lingyu Xu1, Siqi Jiang1, Chenyu Li1,2
1Department of Nephrology, the Affiliated Hospital of Qingdao University, Qingdao, China.
Insights
This study developed a machine learning model to predict acute kidney injury (AKI) and acute kidney disease (AKD) in children. Early detection through this tool can improve outcomes for pediatric patients at risk.
Area of Science:
- Pediatric Nephrology
- Medical Informatics
- Machine Learning in Healthcare
Background:
- Acute kidney injury (AKI) and acute kidney disease (AKD) are common in hospitalized children, increasing mortality and hospital stays.
- Early identification of kidney injury is critical for improving patient outcomes.
- This study addresses the need for timely detection of AKI and AKD in pediatric populations.
Purpose of the Study:
- To develop and validate a machine learning-based risk prediction model for AKI and AKD in pediatric patients.
- To enable personalized risk predictions for early intervention.
- To create an accessible tool for predicting AKI and AKD.
Main Methods:
- Utilized data from 2,346 hospitalized pediatric patients (January 2020 - January 2023).
- Trained and tested eight machine learning and two ensemble algorithms, selecting the optimal model via AUROC.
- Employed SHAP for model interpretability and developed an online prediction tool using Streamlit.
Main Results:
- Incidence rates for AKI and AKD were 14.90% and 16.26%, respectively.
- The LightGBM algorithm demonstrated strong predictive performance (AUROC: 0.813 for AKI, 0.744 for AKD).
- Key predictors identified included serum creatinine for AKI and proton pump inhibitors for AKD.
Conclusions:
- The high incidence of AKI and AKD in children necessitates proactive management.
- Machine learning models, supported by a web-based tool, can effectively predict AKI and AKD in pediatric patients.
- Early identification of high-risk children via these models has the potential to improve clinical outcomes.
Background:
Acute kidney injury (AKI) and acute kidney disease (AKD) are prevalent among pediatric patients, both linked to increased mortality and extended hospital stays. Early detection of kidney injury is crucial for improving outcomes. This study presents a machine learning-based risk prediction model for AKI and AKD in pediatric patients, enabling personalized risk predictions.
Methods:
Data from 2,346 hospitalized pediatric patients, collected between January 2020 and January 2023, were divided into an 85% training set and a 15% test set. Predictive models were constructed using eight machine learning algorithms and two ensemble algorithms, with the optimal model identified through AUROC. SHAP was used to interpret the model, and an online prediction tool was developed with Streamlit to predict AKI and AKD.
Results:
The incidence of AKI and AKD were 14.90% and 16.26%, respectively. Patients with AKD combined with AKI had the highest mortality rate, at 6.94%, when analyzed by renal function trajectories. The LightGBM algorithm showed superior predictive performance for both AKI and AKD (AUROC: 0.813, 0.744). SHAP identified top predictors for AKI as serum creatinine, white blood cell count, neutrophil count, and lactate dehydrogenase, while key predictors for AKD included proton pump inhibitor, blood glucose, hemoglobin, and AKI grade.
Conclusion:
The high incidence of AKI and AKD among hospitalized children warrants attention. Renal function trajectories are strongly associated with prognosis. Supported by a web-based tool, machine learning models can effectively predict AKI and AKD, facilitating early identification of high-risk pediatric patients and potentially improving outcomes.
