OSCR

A machine learning framework for multimodal temporal prediction of neurological outcome after out-of-hospital cardiac arrest.

Overview

Authors: Jee Yong Lim1, Han Joon Kim2
ORCID iDs: Jee Yong Lim
  1. International Healthcare Center, Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, Seoul, Republic of Korea
  2. Department of Emergency Medicine, Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, 222 Banpo-daero, Seocho-gu, Seoul, 06591 Republic of Korea
Institutions: The Catholic University of Korea Seoul St. Mary's Hospital (South Korea); Catholic University of Korea (South Korea)
Journal: Scientific reports, volume 16, issue 1, article 28810
Dates: received 11 March 2026; accepted 17 June 2026; published online 23 June 2026
Type: Research article · Language: English
License: CC BY
Identifiers: DOI 10.1038/s41598-026-59052-2 · PMID 42336956 · PMCID PMC13578319 · OpenAlex W7165614769
Open access: gold, a free copy (OpenAlex)
Status: code on request
Categories: human (organism), clinical / translational (subfield)
Methods: Connectivity, Statistics, Machine learning, Physiology & signal measures
Keywords: Cardiac arrest, Neurological prognostication, Machine learning, Multimodal prediction, Thromboinflammation, Explainable artificial intelligence, Targeted temperature management, Biomarkers, Medical research, Neurology, Neuroscience
MeSH: Machine Learning*, Out-of-Hospital Cardiac Arrest*, Aged, Female, Humans, Hypothermia, Induced, Male, Middle Aged, Predictive Learning Models, Prognosis, Republic of Korea, Retrospective Studies (* major topic)
Topic: Cardiac Arrest and Resuscitation (Emergency Medicine, Medicine), according to OpenAlex
Funding: the Research Fund of Seoul St. Mary's Hospital, The Catholic University of Korea (ZC25RISI0423)
Citations: not cited yet (Europe PMC); 38 references in the paper

Abstract

Neurological prognostication after out-of-hospital cardiac arrest (OHCA) remains challenging. Existing clinical scores rely on static, single-timepoint assessments and fail to capture the dynamic interplay among coagulation derangement, systemic inflammation, brain injury, and evolving neurological status. Whether integrating serial multimodal data through machine learning can meaningfully improve prediction over established approaches has not been systematically evaluated. We conducted a retrospective cohort study of 414 consecutive OHCA patients treated with targeted temperature management (TTM) at a tertiary cardiac arrest center in South Korea (2009–2021), where withdrawal of life-sustaining treatment is not practiced. Ninety-one features spanning five modalities—coagulation, inflammation, brain injury biomarkers, neurological examination, and static clinical variables—were extracted at admission, 24 h, and 48 h. We compared four machine learning algorithms against single-modality models and a clinical score approximation using five-fold stratified cross-validation. Dynamic prediction models evaluated discriminative performance evolution. SHapley Additive exPlanations (SHAP) analysis quantified feature- and modality-level contributions. Robustness was assessed through temporal validation, self-fulfilling prophecy sensitivity analyses, and exclusion of clinician-decision-dependent variables. Of 414 patients (mean age 55.6 years, 71.8% male), 131 (31.6%) achieved favorable neurological outcome (Cerebral Performance Category [CPC] 1–2) at six months. The full multimodal random forest model achieved an area under the receiver operating characteristic curve (AUROC) of 0.983 (95% CI 0.972–0.991), significantly outperforming the clinical score approximation (AUROC 0.847; ΔAUROC + 0.136, p < 0.001) and every single-modality model (all p < 0.001). At 100% specificity, sensitivity was 0.519. Dynamic prediction improved from AUROC 0.950 at admission to 0.977 at 24 h (p < 0.001) and 0.981 at 48 h. SHAP analysis revealed that neurological examination and brain injury biomarkers dominated overall prediction, while coagulation markers—particularly initial international normalized ratio (INR)—provided the strongest early discriminative signal. The model remained robust on temporal validation (AUROC 0.977), after neurological examination exclusion (0.965), and after excluding clinician-decision-dependent variables (0.979). A multimodal machine learning framework integrating serial thromboinflammation, brain injury, and neurological data substantially outperforms conventional approaches for neurological prognostication after OHCA. The dynamic prediction capability and modality-level explainability offer a pathway toward clinically actionable, time-evolving decision support in post-cardiac arrest care.

Supplementary Information: The online version contains supplementary material available at https://doi.org/10.1038/s41598-026-59052-2.

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

Code

The paper says that its authors' code is available on request: it was not published with the paper, so there is nothing to verify.

Code availability

The underlying code for this study is not publicly available but may be made available to qualified researchers on reasonable request from the corresponding author.

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

Tracing map

A tracing map links a paper to the code its authors published: this paper has none (its code is available on request), so it has no map.

Data

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

Data availability

No datasets were generated or analysed during the current study.

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, 2 authors, 11 keywords, 12 MeSH terms, 1 funder, 30 references.

Cite

This paper

Lim, J. Y., & Kim, H. J. (2026). A machine learning framework for multimodal temporal prediction of neurological outcome after out-of-hospital cardiac arrest. Scientific reports, 16(1), 28810. https://doi.org/10.1038/s41598-026-59052-2

BibTeX

@article{lim2026machine,
author = {Lim, Jee Yong and Kim, Han Joon},
title = {{A machine learning framework for multimodal temporal prediction of neurological outcome after out-of-hospital cardiac arrest}},
journal = {Scientific reports},
year = {2026},
month = jun,
volume = {16},
number = {1},
pages = {28810},
publisher = {Nature Publishing Group},
issn = {2045-2322},
doi = {10.1038/s41598-026-59052-2},
url = {https://doi.org/10.1038/s41598-026-59052-2},
pmid = {42336956},
pmcid = {PMC13578319}
}

RIS

TY - JOUR
AU - Lim, Jee Yong
AU - Kim, Han Joon
TI - A machine learning framework for multimodal temporal prediction of neurological outcome after out-of-hospital cardiac arrest
T2 - Scientific reports
J2 - Sci Rep
PY - 2026
DA - 2026/06/23
VL - 16
IS - 1
SP - 28810
SN - 2045-2322
PB - Nature Publishing Group
DO - 10.1038/s41598-026-59052-2
UR - https://doi.org/10.1038/s41598-026-59052-2
LA - en
ER -

CSL-JSON

{
"id": "10.1038/s41598-026-59052-2",
"type": "article-journal",
"title": "A machine learning framework for multimodal temporal prediction of neurological outcome after out-of-hospital cardiac arrest",
"container-title": "Scientific reports",
"author": [
{
"family": "Lim",
"given": "Jee Yong"
},
{
"family": "Kim",
"given": "Han Joon"
}
],
"container-title-short": "Sci Rep",
"volume": "16",
"issue": "1",
"page": "28810",
"DOI": "10.1038/s41598-026-59052-2",
"PMID": "42336956",
"PMCID": "PMC13578319",
"ISSN": "2045-2322",
"publisher": "Nature Publishing Group",
"URL": "https://doi.org/10.1038/s41598-026-59052-2",
"language": "en",
"issued": {
"date-parts": [
[
2026,
6,
23
]
]
}
}

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:10.1186/s13054-026-05929-7
Reporting practices and impact of withdrawal of life-sustaining treatment on outcomes in acute brain injury clinical trials: a literature review and simulation study.
Journal: Critical care (London, England)
In common: clinical / translational, 2 references
[2] doi:10.1093/braincomms/fcag117
Connectome disruptions after hypoxic-ischaemic injury associate with consciousness disorder severity.
Journal: Brain communications
In common: 2 references
[3] doi:10.1038/s41746-026-02708-0 [code]
Multi-dimensional MRI representation and privileged learning approaches to functional outcome prediction for ischemic stroke patients.
Journal: NPJ digital medicine
In common: clinical / translational, 1 reference
[4] doi:10.3390/jcm15176822
Predictors of Six-Month Functional Outcome After Chronic Subdural Hematoma Surgery: Logistic Regression Versus Machine Learning.
Journal: Journal of clinical medicine
In common: clinical / translational, 1 reference
[5] doi:10.3389/fneur.2026.1812084 [code]
Prognostic assessment of comatose patients after cardiac arrest: construction of a simplified monitoring EEG risk score across different time windows.
Journal: Frontiers in neurology
In common: clinical / translational, 1 reference
[6] doi:10.1002/gps3.70039
Associations between wearable-device-measured daytime and nighttime light exposures and dementia risk: A prospective cohort study.
Journal: General psychiatry
In common: clinical / translational, 1 reference
[7] doi:10.1038/s41467-026-73878-4 [code]
Neural Response to Familiar Names Predicts Outcome of Comatose ICU Patients: A Prospective Observational Cohort Study.
Journal: Nature communications
In common: clinical / translational, 1 reference
[8] doi:10.3390/e28070769
Correlation-Induced Accessibility Bridges in Biomedical Networks: A Proof-of-Concept Relational Graph Model.
Journal: Entropy (Basel, Switzerland)
In common: clinical / translational, 1 reference
[9] doi:10.1186/s40478-026-02286-y [code]
AI guided discovery of a murine model of asymptomatic Alzheimer's disease.
Journal: Acta neuropathologica communications
In common: 1 reference
[10] doi:10.1002/alz.71433
Genetic correlation analysis of Alzheimer's disease and stroke implicates PHLPP1 as a shared locus in individuals of African ancestry.
Journal: Alzheimer's & dementia : the journal of the Alzheimer's Association
In common: 1 reference

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.

Sign in with ORCID to claim this paper as one of its authors, correct its record or validate its tracing map: when the paper's metadata lists your ORCID iD, you are recognized at once. Maintainers of its code: sign in with GitHub, then claim the repository on your account page.

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, reproductions, activity

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.