JMIR Publications

ABSTRACT

Background:

Infants born at extremely preterm gestational ages are typically admitted to the neonatal intensive care unit (NICU) after initial resuscitation. The subsequent hospital course can be highly variable, and despite counseling aided by available risk calculators, there are significant challenges with shared decision-making regarding life-support and transition to end-of-life care. Improving predictive models can help providers and families navigate these unique challenges.

Objective:

Machine learning methods have previously demonstrated added predictive value for determining ICU outcomes, and their use allows consideration of a greater number of factors that potentially influence newborn outcomes, such as maternal characteristics. Machine learning-based models were analyzed for their ability to predict survival of extremely preterm neonates at initial admission.

Methods:

Maternal and newborn information was extracted from the health records of infants born between 23 and 29 weeks of gestation in the MIMIC-III critical care database. Applicable machine learning models predicting survival during the initial NICU admission were developed and compared. The same type of models were also examined using only features which would be available prepartum, for the purpose of survival prediction prior to an anticipated preterm birth. Features most correlated with the predicted outcome were determined when possible for each model.

Results:

The resulting random forest model showed higher predictive performance than the frequently utilized SNAPPE-II NICU model when considering extremely preterm infants of very low birth weight. Several other machine learning models were found to have good performance, but did not show a statistical significant difference from previously available models in this study. Feature importance varied by model, and those of greater importance included gestational age, birth weight, initial oxygenation level, elements of the APGAR score, and amount of blood pressure support. Important prepartum features also included maternal age, steroid administration, and presence of pregnancy complications.

Conclusions:

Machine learning methods have potential to provide robust prediction of survival in the context of extremely preterm births and allow for consideration of additional factors such as maternal clinical and socioeconomic information. Evaluation of larger, more diverse datasets may provide additional clarity on comparative performance.