EUVIMEDEUVIMED entdecken
Open Access

Machine Learning Predicts Acute Kidney Injury in Hospitalized Patients with Sickle Cell Disease

Rima S Zahr, Akram Mohammed, Surabhi Naik, Daniel Faradji, Kenneth I Ataga, Jeffrey Lebensburger, Robert L Davis

American Journal of Nephrology · 2024 · 55 (1) · 18-24

DOI: 10.1159/000534864 ↗

Kurz gefasst

Fachartikel in der Zeitschrift American Journal of Nephrology, online erschienen am 31.10.2023. Verfasst von Rima S Zahr und 6 weiteren Autor:innen. Beteiligte Einrichtung: Division of Pediatric Nephrology and Hypertension, University of Tennessee Health Science Center Memphis, Memphis, Tennessee, USA. Hauptthemen laut MEDLINE: Acute Kidney Injury, Anemia, Sickle Cell. Die Arbeit wurde bisher 2-mal zitiert (Crossref). Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NHLBI NIH HHS.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Introduction

Acute kidney injury (AKI) is common among hospitalized patients with sickle cell disease (SCD) and contributes to increased morbidity and mortality. Early identification and management of AKI is essential to preventing poor outcomes. We aimed to predict AKI earlier in patients with SCD using a machine-learning model that utilized continuous minute-by-minute physiological data.

Methods

A total of6,278 adult SCD patient encounters were admitted to inpatient units across five regional hospitals in Memphis, TN, over 3 years, from July 2017 to December 2020. From these, 1,178 patients were selected after filtering for data availability. AKI was identified in 82 (7%) patient encounters, using the 2012 Kidney Disease Improving Global Outcomes (KDIGO) criteria. The remaining 1,096 encounters served as controls. Features derived from five physiological data streams, heart rate, respiratory rate, and blood pressure (systolic, diastolic, and mean), captured every minute from bedside monitors were used. An XGBoost classifier was used for classification.

Results

Our model accurately predicted AKI up to 12 h before onset with an area under the receiver operator curve (AUROC) of 0.91 (95% CI [0.89–0.93]) and up to 48 h before AKI with an AUROC of 0.82 (95% CI [0.80–0.83]). Patients with AKI were more likely to be female (64.6%) and have history of hypertension, pulmonary hypertension, chronic kidney disease, and pneumonia than the control group.

Conclusion

XGBoost accurately predicted AKI as early as 12 h before onset in hospitalized SCD patients and may enable the development of innovative prevention strategies.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Acute Kidney InjuryAnemia, Sickle CellAdultHumansFemaleMaleKidneyRisk AssessmentMachine LearningRetrospective Studies

Stichwörter der Autor:innen

Acute kidney injuryMachine learningPredictive analyticsSickle cell diseaseTime-series data

Autor:innen und Institutionen

  • Division of Pediatric Nephrology and Hypertension, University of Tennessee Health Science Center Memphis, Memphis, Tennessee, USA.
  • Center for Biomedical Informatics, University of Tennessee Health Science Center, Memphis, Tennessee, USA.
  • Department of Surgery, University of Tennessee Health Sciences Center, Memphis, Tennessee, USA.
  • College of Medicine, University of Tennessee Health Sciences Center, Memphis, Tennessee, USA.
  • Center for Sickle Cell Disease, University of Tennessee Health Sciences Center, Memphis, Tennessee, USA.
  • Division of Pediatric Hematology and Oncology, University of Alabama at Birmingham, Birmingham, Alabama, USA.

Förderung

  • NHLBI NIH HHS

Zitieren

APA

Zahr, R. S., Mohammed, A., Naik, S., Faradji, D., Ataga, K. I., Lebensburger, J., & Davis, R. L. (2024). Machine Learning Predicts Acute Kidney Injury in Hospitalized Patients with Sickle Cell Disease. American Journal of Nephrology, 55(1), 18-24. https://doi.org/10.1159/000534864

Vancouver

Zahr RS, Mohammed A, Naik S, Faradji D, Ataga KI, Lebensburger J, et al. Machine Learning Predicts Acute Kidney Injury in Hospitalized Patients with Sickle Cell Disease. American Journal of Nephrology. 2024;55(1):18-24. doi:10.1159/000534864

BibTeX
@article{Zahr2024,
  author = {Zahr, Rima S and Mohammed, Akram and Naik, Surabhi and Faradji, Daniel and Ataga, Kenneth I and Lebensburger, Jeffrey and Davis, Robert L},
  title = {Machine Learning Predicts Acute Kidney Injury in Hospitalized Patients with Sickle Cell Disease},
  journal = {American Journal of Nephrology},
  year = {2024},
  volume = {55},
  number = {1},
  pages = {18--24},
  doi = {10.1159/000534864}
}
Kostenlos starten →