Prediction of neonatal death in pregnant women in an intensive care unit: Application of machine learning models.

Introduction: Neonatal mortality remains a critical concern, particularly in developing countries. The advent of machine learning offers a promising avenue for predicting the survival of at-risk neonates. Further research is required to effectively deploy this approach within distinct clinical contexts. Objective: This study aimed to assess the applicability of machine learning models in predicting neonatal mortality, drawing from maternal and clinical characteristics of pregnant women within an intensive care unit (ICU). Methods: Conducted as an observational cross-sectional study, the research enrolled pregnant women receiving care in a level III national hospital's ICU in Peru. Detailed data encompassing maternal diagnosis, maternal characteristics, obstetric characteristics, and newborn outcomes (survival or demise) were meticulously collected. Employing machine learning, predictive models were developed for neonatal mortality. Estimations of beta coefficients in the training dataset informed the model application to the validation dataset. Results: A cohort of 280 pregnant women in the ICU were included in this study. The Gradient Boosting approach was selected following rigorous experimentation with diverse model types due to its superior F1-score, ROC curve performance, computational efficiency, and learning rate. The final model incorporated variables deemed pertinent to its efficacy, including gestational age, eclampsia, kidney infection, maternal age, previous placenta complications accompanied by hemorrhage, severe preeclampsia, number of prenatal checkups, and history of miscarriages. By incorporating optimized hyperparameter values, the model exhibited an impressive area under the curve (AUC) of 0.98 (95 % CI: 0.95-1), along with a sensitivity of 0.98 (95 % CI: 0.94-1) and specificity of 0.98 (95 % CI: 0.93-1). Conclusion: The findings underscore the utility of machine learning models, specifically Gradient Boosting, in foreseeing neonatal mortality among pregnant women admitted to the ICU, even when confronted with maternal morbidities. This insight can enhance clinical decision-making and ultimately reduce neonatal mortality rates.

Relación entre la luna llena y la incidencia en partos prematuros vaginales en un hospital nivel III de Lima, Perú / Relationship between the full moon and the incidence of premature vaginal births in a level III hospital in Lima, Peru

Objetivo. Determinar la asociación entre la luna llena y la incidencia de partos prematuros vaginales entre mujeres con parto vaginal de un hospital de tercer nivel de Lima, Perú. Material y método. Se realizó un estudio transversal analítico de base secundaria del Certificado de Nacido Vivo (CNV) de Perú. Se estudiaron a todos los recién nacidos del Instituto Nacional Materno Perinatal entre los años 2013 a 2021. La duración de la fase de luna llena se determinó a través de lenguaje de programación con Python 3.6 y el análisis de la incidencia de prematuridad con el paquete estadístico STATA v15. Resultados. Se seleccionaron 90 653 recién nacidos del CNV de los cuales 11563 (12.75%) participantes nacieron durante los días de luna llena y 79089 (87.25%) durante las otras fases. Se observó una mayor incidencia de partos prematuros vaginales durante la fase de luna llena en comparación con otras fases (p<0.01). El análisis multivariado encontró que la luna llena tenía un 1.17% más de valor promedio de incidencia de partos prematuros vaginales ajustado por año en comparación con las demás fases (IC 95% 1.050 - 1.292, p<0.01). Conclusiones. Se encontró una mayor incidencia de partos prematuros vaginales durante la fase de luna llena en la población estudiada. Se deben tomar con cuidado estos resultados debido a que en el análisis se incluyeron los partos inducidos.