Predictive Modeling of Neonatal Mortality Trends Using Machine Learning: A Cross-sectional Data Science Approach

Authors

  • Kaushika Pal
  • Jitendra K Pal
  • Sejal R. Trivedi
  • Neha Sharma
  • Prashant P Patavardhan
  • Sarbojeet Goswami

Keywords:

Neonatal Mortality, Global Health, Mortality Prediction, Data-Driven, Public Health Data

Abstract

In this paper, we propose a machine-learning framework to estimate neonatal mortality rates from cross-sectional global health data. The authors combine cause-specific neonatal death rates and aggregate health statistics from the World Health Organization (WHO) to implement regression models for predicting the total neonatal mortality by constructing a merged dataset spanning multiple countries and years. Univariate analysis shows significant associations between causes such as prematurity, infection and birth trauma with overall mortality. We use multiple machine learning techniques like Linear Regression, Random Forest Regressor, Support Vector Regression, and Gradient Boosting to predict the inferential time of the given queries. Random Forest has the best predictive accuracy among them (R² = 0.990; RMSE = 1.24). These findings validate the feasibility of using machine learning for accurate, data-driven predictions of neonatal mortality to enhance public health policy decisions in relation to maternal, perinatal and neonatal care.

Downloads

Download data is not yet available.

Metrics

Metrics Loading ...

References

Smith J., et al. (2021). Predicting Infant Mortality Using Public Health Data. Journal of Biomedical Informatics, 118, 103805.

Adebayo T., et al. (2020). Deep Learning for Under-5 Mortality Prediction. Health Informatics Journal, 26(2), 1352-1365.

Zhang, Y., Chen, Q., & Wang, X. (2022). Deep learning-based time series modeling for neonatal complications in Chinese hospitals. IEEE Access, 10, 24490–24502.

Patel, R., & Sharma, D. (2019). Machine learning models for predicting maternal and neonatal risk factors in rural India. BMC Pregnancy and Childbirth, 19(1), 280.

Rahman, A., Hossain, M. M., & Uddin, M. J. (2020). Application of Gradient Boosting for predicting infant mortality in Bangladesh. International Journal of Medical Informatics, 135, 104071.

Liu, L., Johnson, H. L., & Walker, N. (2018). Global, regional, and national causes of child mortality using decision tree models. The Lancet Global Health, 6(4), e361–e370.

Singh, A., & Yadav, S. (2021). Socioeconomic predictors of neonatal mortality in India using NFHS-4 data. Social Science & Medicine, 291, 114479.

Ahmed, M. U., et al. (2020). Prediction of neonatal death in Middle Eastern hospitals using artificial neural networks. Health Information Science and Systems, 8, 15.

Chatterjee, K., & Roy, S. (2022). Feature importance in neonatal mortality prediction using XGBoost: A study on NFHS-4 data. Computers in Biology and Medicine, 145, 105451.

Mohan, A., et al. (2019). Dimensionality reduction for child mortality prediction in Africa: A PCA + ML approach. International Journal of Epidemiology, 48(2), 624–633.

Kim, H. J., et al. (2017). Analyzing the causes of neonatal deaths in Korea using machine learning. Journal of Korean Medical Science, 32(12), 1964–1970.

Nasir, M., Rehman, H., & Baig, M. A. (2018). Ensemble learning for predicting neonatal mortality in Pakistani hospitals. Journal of Medical Systems, 42, 96.

Moyo, S., et al. (2020). Predictive modeling of neonatal mortality in Sub-Saharan Africa using logistic regression. BMJ Global Health, 5(6), e002960.

Hoque, M. M., et al. (2021). Deep neural networks for neonatal mortality prediction with UNICEF health data. Expert Systems with Applications, 185, 115593.

Kumar, V., & Sinha, R. (2023). Automated machine learning for neonatal mortality risk assessment in South Asia. Applied Artificial Intelligence, 37(1), 205–222.

United Nations Inter-agency Group for Child Mortality Estimation (UN IGME). (2024). Cause of Death Estimates, Neonatal and Child Mortality by Age Group [Data file]. Retrieved from: https://childmortality.org/

World Bank. (2024). World Development Indicators: Neonatal Mortality Rate (per 1,000 live births) [Data file]. Retrieved from: https://data.worldbank.org/indicator/SH.DYN.NMRT

Downloads

Published

2025-05-13

How to Cite

1.
Pal K, K Pal J, Trivedi SR, Sharma N, P Patavardhan P, Goswami S. Predictive Modeling of Neonatal Mortality Trends Using Machine Learning: A Cross-sectional Data Science Approach. J Neonatal Surg [Internet]. 2025May13 [cited 2025Oct.14];14(22S):1033-9. Available from: https://www.jneonatalsurg.com/index.php/jns/article/view/5766

Issue

Vol. 14 No. 22S (2025): Journal of Neonatal Surgery

Section

Original Article

License

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

You are free to:

  • Adapt — remix, transform, and build upon the material for any purpose, even commercially.

Terms:

  • Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.
  • No additional restrictions — You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.