Integration of Early Physiological Responses Predicts Later Illness Severity in Preterm Infants

• DOI: 10.1126/scitranslmed.3001304
• Publication Date: 2010-09-08T22:28:47Z
• Authors: Suchi Saria; Anand K. Rajani; Jeffrey Gould; Daphne Koller; Anna A. Penn
• Keywords: Apgar score; Gold standard (test)

Physiological data are routinely recorded in intensive care, but their use for rapid assessment of illness severity or long-term morbidity prediction has been limited. We developed a physiological score preterm newborns, akin to an electronic Apgar score, based on standard signals noninvasively admission neonatal care unit. were able accurately and reliably estimate the probability individual infant's risk severe basis noninvasive measurements. This algorithm was with electronically captured time series from first 3 hours life infants (< =34 weeks gestation, birth weight < =2000 g). Extraction integration state-of-the-art machine learning methods produced severity, PhysiScore. PhysiScore validated 138 leave-one-out method prospectively identify at short- morbidity. provided higher accuracy overall (86% sensitive 96% specificity) than other scoring systems, including score. particularly accurate identifying high related specific complications (infection: 90% 100%; cardiopulmonary: 100%). parameters, short-term variability respiratory heart rates, contributed more invasive laboratory studies. Our flexible methodology automated, rapid, measurements can be easily applied range tasks improve patient resource allocation.