Machine learning can help predict patient deterioration, study says

A new study published in Scientific Reports suggests that a machine learning-driven tool could utilize the electronic medical record to predict patient deterioration and help providers prioritize patients.

The study used EMR data and digital tools to create better alerts for cardiac arrest or intensive care unit transfer events. The prediction tool achieved 100 percent sensitivity for predictions of 95 percent, 85 percent and 70 percent risk of deterioration two to eight hours in advance of clinical alerts, according to the July 11 study. 

The study used 18,648 patient records from a major metropolitan hospital in Queensland, Australia. 

Copyright © 2024 Becker's Healthcare. All Rights Reserved. Privacy Policy. Cookie Policy. Linking and Reprinting Policy.

 

Featured Whitepapers

Featured Webinars