Year: 2025
Type: book-chapter
Abstract: Acute kidney injury is a prevalent and severe complication in hospitalized patients, contributing to increased morbidity and mortality. While numerous predictive models exist, they primarily focus on identifying AKI after its onset rather than forecasting it in advance. We developed a labeling algorithm to capture the earliest onset time of in-hospital AKI based on KDIGO 2012 criteria to address this limitation. We retrospectively analyzed 143,512 in-hospital cases from National Health Insurance Service Ilsan Hospital between 2015 and 2021, applying the algorithm to capture the earliest AKI onset time based on serum creatinine and urine output measurements. Our results showed that 31.97% (45,882) of the cases were identified as AKI, with urine output criteria detecting 87.29% of these cases as the earliest onset indicator.
DOI: 10.3233/shti250415
Language: English
Topic: Acute Kidney Injury Research
Subfield: Nephrology
Field: Medicine
Domain: Health Sciences
Open Access status: hybrid
OpenAlex ID: W4410448142