A hybrid neural network can accurately forecast premature babies' individual mortality risk in order to better guide their care, thanks to scientists at Australia's James Cook University (JCU).
JCU's Stephanie Baker said the Neonatal Artificial Intelligence Mortality Score (NAIMS) network assesses preterm infants' mortality risk based on simple demographics and trends in heart and respiratory rate.
Baker said NAIMS could predict an infant's mortality risk within three, seven, or 14 days from data generated over 12 hours, without requiring invasive procedures or knowledge of medical histories.
Said Baker, "Due to the simplicity and high performance of our proposed scheme, NAIMS could easily be continuously and automatically recalculated, enabling analysis of a baby's responsiveness to treatment and other health trends."
From James Cook University (Australia)
View Full Article
Abstracts Copyright © 2021 SmithBucklin, Washington, DC, USA