Artificial intelligence-based model predicts patients' risk of acute kidney injury

Credit: Pixabay/CC0 Public Domain

Acute kidney injury (AKI) is common in patients in intensive care units, and predicting which patients are at risk can help clinicians take appropriate preventive measures. Investigators have recently developed an artificial intelligence-based model to help make such predictions. The research will be presented at ASN Kidney Week 2022 November 3–6.

Among 16,785 adults admitted to the in 2015–2020 in Taichung Veterans General Hospital, 30% developed AKI. An -based AKI prediction model based on these patients' data (21 features including urine trend and serum creatine) was validated in patients from 4 other medical centers (2,874, 10,758, 12,299, and 12,483 patients, respectively, with a wide range of AKI incidence of 24.9–67.2%). The model was accurate at predicting AKI 24 hours ahead of time.

"Early prediction of AKI ahead of 24 hours may help clinicians initiate timely interventions to prevent AKI from happening or alleviate its severity," said corresponding author Chun-Te Huang, MD, of Taichung Veterans General Hospital, in Taiwan. "Our model could be easily shared and integrated to different hospitals to provide a real-time risk prediction in electronic health information systems."

More information: Machine learning for development of a real time AKI risk prediction model in ICU with external validation and federated learning at five medical centers: From model development to clinical application (2022).

Citation: Artificial intelligence-based model predicts patients' risk of acute kidney injury (2022, November 3) retrieved 1 February 2023 from
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without the written permission. The content is provided for information purposes only.

Explore further

Artificial intelligence approach may help identify melanoma survivors who face a high risk of cancer recurrence


Feedback to editors