Abstract

Although prematurity is the single largest cause of death in children under 5 years of age, the current definition of prematurity, based on gestational age, lacks the precision needed for guiding care decisions. Here, we propose a longitudinal risk assessment for adverse neonatal outcomes in newborns based on a deep learning model that uses electronic health records (EHRs) to predict a wide range of outcomes over a period starting shortly before conception and ending months after birth. By linking the EHRs of the Lucile Packard Children’s Hospital and the Stanford Healthcare Adult Hospital, we developed a cohort of 22,104 mother-newborn dyads delivered between 2014 and 2018. Maternal and newborn EHRs were extracted and used to train a multi-input multitask deep learning model, featuring a long short-term memory neural network, to predict 24 different neonatal outcomes. An additional cohort of 10,250 mother-newborn dyads delivered at the same Stanford Hospitals from 2019 to September 2020 was used to validate the model. Areas under the receiver operating characteristic curve at delivery exceeded 0.9 for 10 of the 24 neonatal outcomes considered and were between 0.8 and 0.9 for 7 additional outcomes. Moreover, comprehensive association analysis identified multiple known associations between various maternal and neonatal features and specific neonatal outcomes. This study used linked EHRs from more than 30,000 mother-newborn dyads and would serve as a resource for the investigation and prediction of neonatal outcomes. An interactive website is available for independent investigators to leverage this unique dataset: https://maternal-child-health-associations.shinyapps.io/shiny\_app/. Machine learning models that use maternal and neonatal EHR data can assist in the assessment of risk of multiple adverse neonatal outcomes. Reduction of neonatal mortality and morbidity requires timely risk assessment so that care can be appropriately managed. Using multiple cohorts of mother and newborn dyads, De Francesco et al. trained and externally validated a deep learning model to predict different adverse neonatal outcomes by mining the paired electronic health records (EHRs). Their method largely outperformed currently used EHR-based clinical risk scores and can be applied to EHR data at time points ranging from early gestation to at or after birth, positioning their risk assessment model to be of potential clinical utility. —CAC

Description

impactfactor = {19.343},impactfactor-year = {2023},impactfactor-source = {https://academic-accelerator.com/Impact-of-Journal/Science-Translational-Medicine}

Links and resources

Tags