Development of a machine learning model to identify the predictors of the neonatal intensive care unit admission

Scientists aim to create a system that can predict the likelihood of newborns being admitted to the neonatal intensive care unit (NICU) by combining various statistical methods. This prediction could potentially reduce the negative health outcomes, deaths, and medical costs associated with NICU stays by detecting potential cases early on. This study utilized a retrospective cohort design. The primary outcome of the research focused on admissions to the NICU. The real-time data of pregnant women with a cephalic presentation who gave birth between January 2020 and December 2022 were extracted from the electronic health records of Khaleej-e-Fars Hospital in Bandar Abbas, Iran. The first step of the analysis involved comparing healthy babies to those admitted to the NICU. Variables that had a significant p-value (less than 0.05) were selected as features for the machine learning approach. The input data were utilized to train nine different machine learning models. In our assessment, we used the area under the receiver operating characteristic curve (AUC), accuracy, precision, recall, and F1- score to evaluate the effectiveness. During the study period, the rate of NICU admission at our center was 477 out of 3,062 deliveries (15.5%). In comparison to other models, the random forest classification had the highest accuracy (0.87) and AUC (0.87) for predicting NICU admission. According to our findings, the most significant predictors of NICU admission among several maternal and clinical factors were gestational age, maternal age, parity, a history of neonatal death, onset of labor, multiple pregnancy, fetal distress, meconium-stained amniotic fluid, method of childbirth, neonatal weight, and sex. We have identified several important factors that increase the likelihood of newborns being admitted to the NICU, which could assist in predicting the need for additional neonatal care during delivery and in advising women on the chances of NICU admission.