DeepSurv: A machine learning model for predicting mortality in very-low-birth-weight infants treated in intensive care units

In this study, we developed and evaluated prognostic models for mortality in very-low-birth-weight (VLBW) neonates in the ICU. Our goal was to thoroughly compare the performance of the DeepSurv model with that of traditional machine learning models, such as Cox proportional hazards (CoxPH) and random survival forest (RSF), including temporal metrics.A retrospective analysis of 958 VLBW neonates was performed using data from an integrated health information system from January 2021 to May 2025. The Boruta algorithm, a feature selection method that identifies statistically significant predictors by comparing their importance scores with those of random, shadow variables, identified 11 such predictors in our study.The results showed that DeepSurv demonstrated superior discriminatory ability with an overall AUC of 0.95, significantly outperforming RSF (AUC of 0.89) and CoxPH (AUC of 0.86). The temporal AUC analysis also favored DeepSurv. Moreover, DeepSurv recorded the lowest Brier scores (IBS 0.071) throughout the entire follow-up period compared with CoxPH (IBS 0.129) and RSF (IBS 0.126). Calibration curves confirmed the high predictive accuracy of all the models, and Kaplan–Meier analysis along with log-rank tests demonstrated their effectiveness in risk stratification. Decision curve analysis revealed that the DeepSurv model had the highest clinical utility at both day 7 and day 28, underscoring its potential impact on neonatal care. This study confirms the superiority of DeepSurv in predicting neonatal mortality, providing a valuable tool for risk stratification in clinical practice.