Related Experiment Video
Updated: Oct 28, 2025
Modeling Ascending Vaginal Infection, Preterm Birth, and Neonatal Morbidity in Mice
Published on: October 10, 2025
Machine Learning Models for Predicting Neonatal Mortality: A Systematic Review
Cheyenne Mangold1, Sarah Zoretic2, Keerthi Thallapureddy1
1Department of Pediatrics, University of Texas Health San Antonio, San Antonio, Texas, USA.
Insights
Artificial intelligence (AI) and machine learning (ML) models show promise in accurately predicting neonatal mortality. Future research should prioritize external validation and calibration for clinical application.
Area of Science:
- Medical Informatics
- Computational Biology
- Public Health
Background:
- Neonatal mortality remains a critical global health challenge, with nearly half of all under-five child deaths occurring in the first month of life.
- Identifying neonates at high risk is crucial for targeted interventions and improving global child survival rates.
- Artificial intelligence (AI) offers potential for early identification of modifiable risk factors in neonatal mortality.
Purpose of the Study:
- To systematically review and analyze studies utilizing AI for predicting neonatal mortality.
- To identify commonly used AI models, predictors, and performance metrics in neonatal mortality prediction.
- To assess the quality of evidence regarding AI-driven neonatal mortality prediction.
Main Methods:
- A comprehensive literature search was conducted across major databases (PubMed, Cochrane, OVID, Google Scholar).
- Studies employing AI, including machine learning (ML) and deep learning, for neonatal death prediction were included; studies with <500 participants or using only antenatal factors were excluded.
- Data extraction focused on study design, AI models, features, validation methods, and performance metrics (AUC, sensitivity, specificity).
Main Results:
- Eleven studies involving 1.26 million neonates were included, with predictions made from 5 minutes to 7 days of life.
- Neural networks, random forests, and logistic regression were common AI models; Area Under the Curve (AUC) ranged from 58.3% to 97.0%.
- While models showed varying accuracy (sensitivity 63-80%, specificity 78-99%), only a minority reported external validation or calibration.
Conclusions:
- AI, particularly ML models, demonstrates significant potential for accurate prediction of neonatal mortality.
- This review highlights prevalent AI predictors and evaluation metrics, informing future model development.
- Further research must emphasize external validation, robust calibration, and the development of accessible clinical applications for AI-driven neonatal care.
Introduction:
Approximately 7,000 newborns die every day, accounting for almost half of child deaths under 5 years of age. Deciphering which neonates are at increased risk for mortality can have an important global impact. As such, integrating high computational technology (e.g., artificial intelligence [AI]) may help identify the early and potentially modifiable predictors of neonatal mortality. Therefore, the objective of this study was to collate, critically appraise, and analyze neonatal prediction studies that included AI.
Methods:
A literature search was performed in PubMed, Cochrane, OVID, and Google Scholar. We included studies that used AI (e.g., machine learning (ML) and deep learning) to formulate prediction models for neonatal death. We excluded small studies (n < 500 individuals) and studies using only antenatal factors to predict mortality. Two independent investigators screened all articles for inclusion. The data collection consisted of study design, number of models, features used per model, feature importance, internal and/or external validation, and calibration analysis. Our primary outcome was the average area under the receiving characteristic curve (AUC) or sensitivity and specificity for all models included in each study.
Results:
Of 434 articles, 11 studies were included. The total number of participants was 1.26 M with gestational ages ranging from 22 weeks to term. Number of features ranged from 3 to 66 with timing of prediction as early as 5 min of life to a maximum of 7 days of age. The average number of models per study was 4, with neural network, random forest, and logistic regression comprising the most used models (58.3%). Five studies (45.5%) reported calibration plots and 2 (18.2%) conducted external validation. Eight studies reported results by AUC and 5 studies reported the sensitivity and specificity. The AUC varied from 58.3% to 97.0%. The mean sensitivities ranged from 63% to 80% and specificities from 78% to 99%. The best overall model was linear discriminant analysis, but it also had a high number of features (n = 17).
Discussion/Conclusion:
ML models can accurately predict death in neonates. This analysis demonstrates the most commonly used predictors and metrics for AI prediction models for neonatal mortality. Future studies should focus on external validation, calibration, as well as deployment of applications that can be readily accessible to health-care providers.
Related Concept Videos
01:23Mechanistic Models: Compartment Models in Individual and Population Analysis
01:29Mechanistic Models: Compartment Models in Algorithms for Numerical Problem Solving
In individual population analyses, different algorithms are employed, such as Cauchy's method, which uses a...