Thursday , June 24 2021

The model predicts acute kidney injury requiring dialysis in patients with COVID-19



kidney

CC0 credit: public domain

A new algorithm based on artificial intelligence can help doctors predict which patients with COVID-19 face a high risk of developing acute kidney injury (AKI) that requires dialysis. The survey will be presented online during ASN Kidney Week 2020 Reimagined from October 19 to October 25.

Preliminary reports indicate that acute AKI is common in patients with COVID-19. Using data from more than 3,000 patients hospitalized with COVID-19, researchers at the Icahn School of Medicine on Mount Sinai trained a model based on machine learning, a type of artificial intelligence, to predict ARI that requires dialysis. Only the information collected in the first 48 hours after admission was included, so that predictions could be made when patients were admitted.

The model demonstrated high accuracy (AUC of 0.79), and important characteristics for the prediction included blood levels of creatinine and potassium, age and vital signs of heart rate and oxygen saturation.

“A machine learning model using admission resources has performed well to predict the need for dialysis. Models like this are potentially useful for resource allocation and planning during future outbreaks of COVID-19,” said co-author Lili Chan, MD, MS. “We are in the process of implementing this model in our health systems to help doctors take better care of their patients.”


Follow the latest news on the coronavirus outbreak (COVID-19)


More information:
Study: “Machine learning for predication of severe acute kidney injury in patients hospitalized with COVID-19”

Provided by the American Society of Nephrology

Quote: The model predicts acute kidney injury requiring dialysis in patients with COVID-19 (2020, October 24) recovered on October 24, 2020 at https://medicalxpress.com/news/2020-10-acute-kidney-injury -requiring-dialysis.html

This document is subject to copyright. In addition to any fair dealing for the purpose of study or private research, no part may be reproduced without written permission. The content is provided for informational purposes only.




Source link