Early prediction of moderate-to-severe BPD or death in preterm infants using minute-level respiratory monitoring and explainable machine learning
Abstract
Background: Early identification of extremely preterm infants at risk of moderate-to-severe bronchopulmonary dysplasia (BPD) or death remains challenging. We assessed whether minute-by-minute respiratory data from the first 24 h improves early prediction of this outcome using explainable machine learning. Methods: Retrospective cohort of 192 infants <30 weeks' gestation (2019-2024). Logistic regression, random forest, and XGBoost models were trained on 17 predictors and summarized 24-h respiratory data. Evaluation used AUROC, AUPRC, Brier score, calibration, and decision curves, with SHAP for interpretability. Results: Forty-five infants (23.4%) reached the composite endpoint. All models achieved comparable discrimination (AUROC 0.831-0.847). Random forest was selected as the headline model based on superior calibration (slope 1.016; ECE 0.053), lowest Brier score (0.131), and highest net benefit at decision thresholds ≥20%. Chest X-ray density, gestational age, and maximum FiO₂ were the top SHAP predictors in the headline model; gestational age achieved the highest cross-model rank. SHAP identified a gestational age threshold at 28 weeks, an FiO₂ risk step at 30-35%, and a U-shaped birthweight z-score effect. Conclusion: Combining early radiographic and high-frequency respiratory data with explainable machine learning enables well-calibrated BPD risk stratification on day 0-1. SHAP decomposition supports personalized care by distinguishing modifiable from non-modifiable risks. Impact: A machine learning model integrating respiratory monitoring, radiography, and perinatal variables from the first 24 h of life predicts moderate-to-severe BPD or death in very preterm infants with an AUROC of 0.831 and well-calibrated risk estimates (calibration slope 1.016). Unlike existing tools relying on data from day 7 or later, this day-1 approach significantly improves early prognosis beyond gestational age and birthweight alone. Providing net clinical benefit across a 10-30% threshold, it enables individualized, actionable risk stratification at the bedside, while clinical trajectories can still be modified.
Related
- An Explainable Hybrid Machine Learning Approach Based on MPSO-XGBoost for Pest Density Prediction in Almond Orchards
- Explainable plant disease classification and severity assessment using MobileNetV3-large for crop health monitoring
- Person‐Reported and Continuous Glucose Monitoring ( CGM ) Recorded Hypoglycaemia in People With Insulin‐Treated Type 2 Diabetes and Mild to Moderate Chronic Kidney Disease; Post Hoc Analysis From the Hypo‐ METRICS Study
- From prediction to actionable mechanisms: Explainable multi‑omics AI for farm‑to‑fork postharvest preservation
- Physics-guided explainable machine learning for predicting temporal scour evolution around hybrid spur dikes under unsteady flood hydrographs
BibTeX
@article{louifi2026early,
title = {{Early prediction of moderate-to-severe BPD or death in preterm infants using minute-level respiratory monitoring and explainable machine learning}},
author = {Théolyne Louifi and Boris Delange and Mona Leroux and Patrick Pladys and Alain Beuchée},
journal = {Pediatric Research},
year = {2026},
month = oct,
doi = {10.1038/s41390-026-05503-5},
url = {https://hal.science/hal-05785922v1},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Early prediction of moderate-to-severe BPD or death in preterm infants using minute-level respiratory monitoring and explainable machine learning
AU - Théolyne Louifi
AU - Boris Delange
AU - Mona Leroux
AU - Patrick Pladys
AU - Alain Beuchée
PY - 2026
DA - 2026/10/08
JO - Pediatric Research
DO - 10.1038/s41390-026-05503-5
UR - https://hal.science/hal-05785922v1
AB - Background: Early identification of extremely preterm infants at risk of moderate-to-severe bronchopulmonary dysplasia (BPD) or death remains challenging. We assessed whether minute-by-minute respiratory data from the first 24 h improves early prediction of this outcome using explainable machine learning. Methods: Retrospective cohort of 192 infants <30 weeks' gestation (2019-2024). Logistic regression, random forest, and XGBoost models were trained on 17 predictors and summarized 24-h respiratory data. Evaluation used AUROC, AUPRC, Brier score, calibration, and decision curves, with SHAP for interpretability. Results: Forty-five infants (23.4%) reached the composite endpoint. All models achieved comparable discrimination (AUROC 0.831-0.847). Random forest was selected as the headline model based on superior calibration (slope 1.016; ECE 0.053), lowest Brier score (0.131), and highest net benefit at decision thresholds ≥20%. Chest X-ray density, gestational age, and maximum FiO₂ were the top SHAP predictors in the headline model; gestational age achieved the highest cross-model rank. SHAP identified a gestational age threshold at 28 weeks, an FiO₂ risk step at 30-35%, and a U-shaped birthweight z-score effect. Conclusion: Combining early radiographic and high-frequency respiratory data with explainable machine learning enables well-calibrated BPD risk stratification on day 0-1. SHAP decomposition supports personalized care by distinguishing modifiable from non-modifiable risks. Impact: A machine learning model integrating respiratory monitoring, radiography, and perinatal variables from the first 24 h of life predicts moderate-to-severe BPD or death in very preterm infants with an AUROC of 0.831 and well-calibrated risk estimates (calibration slope 1.016). Unlike existing tools relying on data from day 7 or later, this day-1 approach significantly improves early prognosis beyond gestational age and birthweight alone. Providing net clinical benefit across a 10-30% threshold, it enables individualized, actionable risk stratification at the bedside, while clinical trajectories can still be modified.
SN - 0031-3998
SN - 1530-0447
ER -