Development of a neural network model for early detection of creatinine change in critically Ill children
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.
Authors
- Celeste G Dixon
- Murray M. Pollack (ORCID: https://orcid.org/0000-0003-0966-8261)
- Anita Patel (ORCID: https://orcid.org/0000-0002-3791-7379)
- Eduardo A. Trujillo Rivera
Institutions
- George Washington University Hospital (US)
- Children's National (US)
- George Washington University (US)
Publication Details
- Journal
- Frontiers in Pediatrics
- Published
- 2025-04-04
- DOI
- https://doi.org/10.3389/fped.2025.1549836
- Citations
- 1
- Primary Topic
- Acute Kidney Injury Research
- Type
- article
- Field-Weighted Citation Impact
- 1.43