Neonatal mortality prediction under extreme class imbalance: calibrated supervised models and exploratory phenotype analysis in a population-based Iranian registry
Abstract
Neonatal mortality remains a leading contributor to under-five deaths, yet early, individualized risk prediction is hampered by the rarity of the outcome and by limited attention to calibration and interpretability in prior models. We developed and stress-tested machine learning (ML) models for neonatal mortality prediction in a large, population-based cohort, integrating probability calibration, explainable AI, and exploratory unsupervised phenotyping. The development cohort of 61,517 neonates (131 deaths; class imbalance 1:469) was used to train L2-regularized logistic regression (LR) and LightGBM models on 60 routinely available perinatal features. performance was assessed by discrimination (AUROC, AUPRC), calibration, and decision-curve analysis, while SHAP and four unsupervised methods (UMAP, K-means, Isolation Forest, Apriori) characterized the underlying risk structure. Locked models then underwent a one-time case-only external stress test on 604 deceased neonates from a separate center and later period. LR achieved an AUROC of 0.955 (LightGBM, 0.944); at the optimal threshold (τ*=0.004) it captured 87.8% of deaths at 97.3% specificity, and positive net clinical benefit. In the external stress test, LR flagged 69.5% of deaths. Exploratory unsupervised analyses, fitted without outcome labels and evaluated post hoc against mortality, described a single high-risk phenotype; a K-means cluster comprising 6.7% of neonates contained 89.3% of all deaths. In this large, heterogeneous cohort, well-regularized ML models provided calibrated, interpretable, and early neonatal mortality risk stratification. The case-only design of the external cohort permits assessment of sensitivity only; external discrimination and calibration remain to be established, and prospective evaluation is warranted before clinical deployment.
Related
- Competing Risk Prediction of COVID-19 Hospital Mortality with Model Diagnostics and Exploratory Evaluation in Mexico
- Stage Migration, Treatment Patterns, and Machine Learning-Based Prediction of 12-Month Mortality in Lung Adenocarcinoma Across the COVID-19 Era: A SEER Population-Based Study
- Interpretable and Calibrated Classification of Clinical Data Using Supervised Feature Binarization
- Incremental value of bilateral ankle-brachial index for ten-year heart and cerebrovascular mortality prediction: an exploratory temporal benchmark in NHANES
- Cross-tissue graph attention networks for semi-supervised gene expression prediction
BibTeX
@article{alamdarloo2026neonatal,
title = {{Neonatal mortality prediction under extreme class imbalance: calibrated supervised models and exploratory phenotype analysis in a population-based Iranian registry}},
author = {Shaghayegh Moradi Alamdarloo and Amirreza Kanafchian and Hamideh Barzegar and Homeira Vafaei Cisakht and Nasrin Asadi and Fatemeh Asghari and Alireza Montaseri and Reza Bahrami and Khadijeh Bazrafshan and Atefe Hashemi and Ali Haghighi},
journal = {Scientific Reports},
year = {2026},
month = oct,
doi = {10.1038/s41598-026-76979-8},
url = {https://www.nature.com/articles/s41598-026-76979-8},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Neonatal mortality prediction under extreme class imbalance: calibrated supervised models and exploratory phenotype analysis in a population-based Iranian registry
AU - Shaghayegh Moradi Alamdarloo
AU - Amirreza Kanafchian
AU - Hamideh Barzegar
AU - Homeira Vafaei Cisakht
AU - Nasrin Asadi
AU - Fatemeh Asghari
AU - Alireza Montaseri
AU - Reza Bahrami
AU - Khadijeh Bazrafshan
AU - Atefe Hashemi
AU - Ali Haghighi
PY - 2026
DA - 2026/10/20
JO - Scientific Reports
DO - 10.1038/s41598-026-76979-8
UR - https://www.nature.com/articles/s41598-026-76979-8
AB - Neonatal mortality remains a leading contributor to under-five deaths, yet early, individualized risk prediction is hampered by the rarity of the outcome and by limited attention to calibration and interpretability in prior models. We developed and stress-tested machine learning (ML) models for neonatal mortality prediction in a large, population-based cohort, integrating probability calibration, explainable AI, and exploratory unsupervised phenotyping. The development cohort of 61,517 neonates (131 deaths; class imbalance 1:469) was used to train L2-regularized logistic regression (LR) and LightGBM models on 60 routinely available perinatal features. performance was assessed by discrimination (AUROC, AUPRC), calibration, and decision-curve analysis, while SHAP and four unsupervised methods (UMAP, K-means, Isolation Forest, Apriori) characterized the underlying risk structure. Locked models then underwent a one-time case-only external stress test on 604 deceased neonates from a separate center and later period. LR achieved an AUROC of 0.955 (LightGBM, 0.944); at the optimal threshold (τ*=0.004) it captured 87.8% of deaths at 97.3% specificity, and positive net clinical benefit. In the external stress test, LR flagged 69.5% of deaths. Exploratory unsupervised analyses, fitted without outcome labels and evaluated post hoc against mortality, described a single high-risk phenotype; a K-means cluster comprising 6.7% of neonates contained 89.3% of all deaths. In this large, heterogeneous cohort, well-regularized ML models provided calibrated, interpretable, and early neonatal mortality risk stratification. The case-only design of the external cohort permits assessment of sensitivity only; external discrimination and calibration remain to be established, and prospective evaluation is warranted before clinical deployment.
SN - 2045-2322
ER -