Related Experiment Video
Updated: Jul 12, 2025

A Large Animal Model for Acute Kidney Injury by Temporary Bilateral Renal Artery Occlusion
Published on: February 2, 2021
Machine Learning Predicts Acute Kidney Injury in Hospitalized Patients with Sickle Cell Disease
Rima S Zahr1, Akram Mohammed2, Surabhi Naik3
1Division of Pediatric Nephrology and Hypertension, University of Tennessee Health Science Center Memphis, Memphis, Tennessee, USA.
This study developed a machine-learning model to predict acute kidney injury (AKI) in hospitalized sickle cell disease (SCD) patients. The model achieved high accuracy, enabling early detection and potential prevention of AKI.
Area of Science:
- Nephrology
- Hematology
- Data Science
Background:
- Acute kidney injury (AKI) is a significant complication in hospitalized sickle cell disease (SCD) patients, increasing morbidity and mortality.
- Early identification and management of AKI are crucial for improving patient outcomes.
Purpose of the Study:
- To develop and validate a machine-learning model for early prediction of AKI in hospitalized SCD patients.
- To utilize continuous minute-by-minute physiological data for enhanced predictive accuracy.
Main Methods:
- A retrospective analysis of 1,178 adult SCD patient encounters from five hospitals.
- AKI identification based on the 2012 Kidney Disease Improving Global Outcomes (KDIGO) criteria.
- An XGBoost classifier trained on minute-by-minute heart rate, respiratory rate, and blood pressure data.
Main Results:
- The XGBoost model accurately predicted AKI up to 12 hours before onset with an AUROC of 0.91.
- The model also predicted AKI up to 48 hours before onset with an AUROC of 0.82.
- AKI patients were more likely to be female and have comorbidities like hypertension and pneumonia.
Conclusions:
- Machine learning, specifically XGBoost, can accurately predict AKI in hospitalized SCD patients.
- Early AKI prediction up to 48 hours in advance may facilitate the development of novel prevention strategies.
Related Concept Videos
Acute Kidney Injury IV: Diagnostic Studies and Prevention
Acute Kidney Injury I: Introduction
Acute Kidney Injury II: Pathophysiology
Acute Kidney Injury VI: Nursing Management
Acute Kidney Injury V: Interprofessional Care
Acute Kidney Injury III: Clinical Manifestations

