Developing and Validating a Machine Learning Model to Predict Brain Injury in Preterm Infants Using Multisource Data from the Early Postnatal Period

Overview

1. Capital Institute of Pediatrics, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing 100020, China; (P.X.); (P.Z.); (Q.L.)
2. Department of Neonatology, Capital Center for Children’s Health, Capital Medical University, Capital Institute of Pediatrics, Beijing 100020, China; (Y.L.); (Y.C.); (T.H.); (D.Z.); (J.C.)

Abstract

Background: Moderate-to-severe preterm brain injury (PBI), including intraventricular hemorrhage (IVH) and periventricular leukomalacia (PVL), remains an important cause of adverse neurodevelopmental outcomes in preterm infants. Early risk stratification using routinely collected clinical data may help prioritize surveillance in vulnerable infants. Methods: We retrospectively included 318 preterm infants admitted between 2015 and 2024 as the development cohort. Thirty-three candidate predictors derived from perinatal factors, first laboratory tests within 24 h of admission, and selected early hospitalization variables were evaluated. Seven machine-learning algorithms were developed using stratified 10 × 5 nested cross-validation with prespecified preprocessing, class-balancing, and feature-selection procedures. Candidate models were compared primarily using the mean fold-level area under the receiver operating characteristic curve (AUROC). After model selection, the finalized LightGBM model was calibrated using Platt scaling, and its pooled out-of-fold (OOF) performance was summarized. Two prespecified thresholds (Youden and high-sensitivity) were used for risk stratification. A small independent temporal cohort of 35 infants was used for preliminary external validation. Results: PBI occurred in 62/318 infants (19.5%) in the development cohort and 6/35 infants (17.1%) in the temporal external cohort. During candidate-model comparison, LightGBM achieved the highest mean fold-level AUROC (0.768, 95% CI 0.708–0.825). The finalized 14-feature LightGBM model, evaluated using pooled OOF predictions after Platt calibration, yielded an AUROC of 0.747 (95% CI 0.679–0.811), a PR-AUC of 0.392, and a Brier score of 0.136. At the Youden threshold (0.18), sensitivity was approximately 0.70 and specificity approximately 0.85; at the high-sensitivity threshold (0.10), sensitivity was approximately 0.95 and specificity approximately 0.50. Key predictors included ventilation status and early physiologic and laboratory indicators. In the small temporal external cohort (n = 35), the AUROC was 0.897 (95% CI 0.672–1.000); however, this high point estimate should not be overinterpreted because of the limited sample size, wide confidence interval, and suboptimal calibration, and should therefore be considered preliminary. Conclusions: We developed an interpretable LightGBM model using routinely available early postnatal and early hospitalization data to support risk stratification for PBI in preterm infants. The model showed moderate internal discrimination and a positive net benefit across clinically relevant thresholds. Preliminary temporal external validation in a small cohort yielded highly uncertain estimates; larger multicenter studies are needed to confirm generalizability, refine calibration, and determine the most appropriate implementation strategy before routine clinical use.

Reproduced under the paper's license (CC BY), from the paper cited above.

Repository

Its files are read in the Code ↔ Paper reader above, with 2 matches between paragraphs and lines of code.

The paper's code and data availability statement is in the Data section.

Tracing map

Proposed by the machine: these links were found in the paper and verified at the source, without human review. The map will receive a Zenodo DOI once one of the paper's authors has validated it with their ORCID.

What the map holds:

• 1 repository of the authors' code, each at its verified commit, with its license and how the link was found in the paper;
• 1 script, each with its path and the digest of its content;
• 2 matches between paragraphs of the paper and lines of the code (method lexical-v1);
• neither the text of the paper nor the code itself.

Its JSON (tracing-map.json) is deposited on Zenodo with its DOI once the map is validated.

Data

No dataset and no data link were found in the paper.

Data Availability Statement

The clinical data are not publicly available because of privacy and institutional restrictions. Researchers interested in the data for academic purposes may contact the corresponding author. The code for the research-use web application is available at https://github.com/lyrperciver/PBI-APP (20 November 2025).

Reproduced under the paper's license (CC BY), from the paper cited above.

Versions

The history of this record: each version stored by the harvester or made by a correction of its authors or of the maintainers of its code, and what changed in its facts. The texts of the paper (its abstract, its availability statements) are not part of it; versions that changed only those are not listed.

Version 1, 27 September 2026: the first record

Recorded: type, language, journal, volume, issue, pages, dates, 9 authors, 5 keywords, 2 funders, 31 references.

Cite

This paper

Xu, P., Li, Y., Chen, Y., Han, T., Zou, P., Lu, Q., Zhang, D., Chen, J., & Wang, Y. (2026). Developing and Validating a Machine Learning Model to Predict Brain Injury in Preterm Infants Using Multisource Data from the Early Postnatal Period. Children (Basel, Switzerland), 13(6), 796.

The tracing map gets a citation of its own once an author has validated it and it has a DOI.

Similar papers

The papers with a page that share the most with this one: the tools found in their code, their categories, datasets, cited references and authors, the rarest counting most.

[1] doi:10.1016/j.isci.2026.117012 [code]

Multi-level static and dynamic graph-theoretical analyses of resting-state functional networks in a Chinese cohort of preterm neonates.

Journal: iScience

In common: scikit-learn, pandas, Matplotlib, 1 other tool, developmental, other condition, 1 reference

[2] doi:10.1162/imag.a.1347 [code]

Neural and behavioural correlates of theory of mind reasoning in five-year-old children born preterm.

Journal: Imaging neuroscience (Cambridge, Mass.)

In common: scikit-learn, pandas, Matplotlib, 1 other tool, other condition, 1 reference

[3] doi:10.1016/j.dcn.2026.101775 [code]

Neonatal brain-age models in full- and preterm infants.

Journal: Developmental cognitive neuroscience

In common: scikit-learn, pandas, Matplotlib, 1 other tool, developmental, other condition

[4] doi:10.1038/s41467-026-73770-1 [code]

Non-coding structural variants disrupt FOXG1 transcriptional regulation in early neurodevelopment.

Journal: Nature communications

In common: scikit-learn, pandas, Matplotlib, 1 other tool, developmental, other condition

[5] doi:10.1186/s40708-026-00312-2 [code]

Synergistic and redundant information dynamics exhibit dissociable alterations across schizophrenia and neurodevelopmental conditions.

Journal: Brain informatics

In common: scikit-learn, pandas, Matplotlib, 1 other tool, developmental, other condition

[6] doi:10.1038/s42003-026-09908-0 [code]

Offspring genetic diversity regulates rearing experiences that predict differential susceptibility to Chd8 haploinsufficiency.

Journal: Communications biology

In common: scikit-learn, pandas, Matplotlib, 1 other tool, developmental, other condition

[7] doi:10.1038/s41467-026-76675-1 [code]

Long-read proteogenomic atlas of human neuronal differentiation reveals isoform diversity informing neurodevelopmental risk mechanisms.

Journal: Nature communications

In common: scikit-learn, pandas, Matplotlib, 1 other tool, developmental

[8] doi:10.21203/rs.3.rs-10335768/v1 [code]

Interneurons in the subplate are associated with cortical maturation across the foetal human cortex

Journal: Research Square (preprint)

In common: scikit-learn, pandas, Matplotlib, 1 other tool, developmental

[9] doi:10.1038/s41467-026-75722-1 [code]

Single-nucleus analysis of the adult human olfactory epithelium uncovers shared neurogenesis programs with the brain.

Journal: Nature communications

In common: scikit-learn, pandas, Matplotlib, 1 other tool, developmental

[10] doi:10.7554/elife.107088 [code]

Development of auditory and spontaneous movement responses to music over the first postnatal year.

Journal: eLife

In common: scikit-learn, pandas, Matplotlib, 1 other tool, developmental

Contribute

The authors of this paper can claim it, correct its record and validate its tracing map, and the maintainers of its code (its owner, or a public member of its organization) correct what it says of their repository; anyone signed in can ask for its removal. Every request goes to OSCR's own machine, which answers it; your account page follows them.

Request its removal

To ask OSCR to remove this record, the copies of its authors' scripts or its tracing map, use the removal request page: signed in, you say who you are, what to remove and why, then review and confirm the request. Published rules decide every request (how).

Discussion: questions and error reports about this paper and its code, from signed-in readers and its authors. It opens with sign-in.

Reproductions: reports from readers who ran the authors' code: what they reproduced, with which environment, commit and data. It opens with sign-in.

Activity: what happens around this paper: new versions of its record, its map's validation, discussions and reproductions. It opens with sign-in.