Machine learning-based scoring model for predicting mortality in ICU-admitted ischemic stroke patients with moderate to severe consciousness disorders - Original research

Aug 5, 2026

Upload by Rowan

Stroke is a leading cause of death and disability, and ischemic stroke patients with moderate to severe consciousness disorders represent a high-risk ICU subgroup requiring dependable outcome guidance. This retrospective study builds and validates an automated machine-learning scoring system to estimate short-term mortality (3 and 7 days) and longer-term mortality (30 and 90 days) using MIMICIV ICU data. Patients with Glasgow Coma Scale criteria indicating moderate to severe impairment were included, and top predictors were selected via an AutoScore framework.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages14

Reading Time35 min

AccessFree to download

Background

Methods

Data source and study population

Candidate variables and model development

Model evaluation

Results

Conclusion

Introduction

Global stroke burden and risk

Definition of consciousness disorders

Clinical decision-making needs and limitations

TYPE Original Research PUBLISHED 18 March 2025 DOI 10.3389/fneur.2025.1534961 OPEN ACCESS EDITED BY Wen-Jun Tu, Capital Medical University, China REVIEWED BY Zhong Wang, The First Affiliated Hospital of Soochow University, China Yong Yang, Guangdong Provincial People’s Hospital, China Hedong Han, Nanjing University, China *CORRESPONDENCE Lei Ye [yelei@ahmu.edu.cn](yelei@ahmu.edu.cn)[ ](yelei@ahmu.edu.cn)Hong-Wei Cheng [hongwei.cheng@ahmu.edu.cn](hongwei.cheng@ahmu.edu.cn)[ ](hongwei.cheng@ahmu.edu.cn)RECEIVED 27 November 2024 ACCEPTED 04 March 2025 PUBLISHED 18 March 2025 CITATION Zhou Z, Chen B, Mei Z-J, Chen W, Cao W, Xu E-X, Wang J, Ye L and Cheng H-W (2025) Machine learning-based scoring model for predicting mortality in ICU-admitted ischemic stroke patients with moderate to severe consciousness disorders. Front. Neurol. 16:1534961 . doi: 10.3389/fneur.2025.1534961 COPYRIGHT © 2025 Zhou, Chen, Mei, Chen, Cao, Xu, Wang, Ye and Cheng. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Machine learning-based scoring model for predicting mortality in ICU-admitted ischemic stroke patients with moderate to severe consciousness disorders Zhou Zhou 1,2, Bo Chen 2, Zhao-Jun Mei 1,2, Wei Chen 1,2, Wei Cao3, En-Xi Xu 2, Jun Wang 2, Lei Ye 1* and Hong-Wei Cheng 1* 1 Department of Neurosurgery, The First Affiliated Hospital of Anhui Medical University, Hefei, China, 2 Department of Neurosurgery, Affiliated People’s Hospital of Jiangsu University, Jiangsu, China, 3 Department of Neurology, Affiliated People’s Hospital of Jiangsu University, Jiangsu, China Background: Stroke is a leading cause of mortality and disability globally. Among ischemic stroke patients, those with moderate to severe consciousness disorders constitute a particularly high-risk subgroup. Accurate predictive models are essential for guiding clinical decisions in this population. This study aimed to develop and validate an automated scoring system using machine learning algorithms for predicting short-term (3-and 7-day) and relatively longterm (30-and 90-day) mortality in this population. Methods: This retrospective observational study utilized data from the MIMICIV database, including 648 ischemic stroke patients with Glasgow Coma Scale (GCS) scores ≤12, admitted to the ICU between 2008 and 2019. Patients with GCS scores indicating speech dysfunction but clear consciousness were excluded. A total of 47 candidate variables were evaluated, and the top six predictors for each mortality model were identified using the AutoScore framework. Model performance was assessed using the area under the curve (AUC) from receiver operating characteristic (ROC) analyses. Results: The median age of the cohort was 76.8 years (IQR, 64.97–86.34), with mortality rates of 8.02% at 3 days, 18.67% at 7 days, 33.49% at 30 days, and 38. 89% at 90 days. The AUCs for the test cohort’s 3-, 7-, 30-, and 90-day mortality prediction models were 0.698, 0.678, 0.724, and 0.730, respectively. Conclusion: We developed and validated a novel machine learning-based scoring tool that effectively predicts both short-term and relatively long-term mortality in ischemic stroke patients with moderate to severe consciousness disorders. This tool has the potential to enhance clinical decision-making and resource allocation for these patients in the ICU. KEYWORDS ischemic stroke, consciousness disorders, scoring model, machine learning, mortality Frontiers in Neurology 01 [frontiersin.org](frontiersin.org) Introduction Stroke, including both ische

FAQs

What patient group does the scoring model target?

The model targets ICU-admitted ischemic stroke patients with moderate to severe consciousness disorders at admission, defined by Glasgow Coma Scale (GCS) scores ≤ 12, with specific exclusions for speech dysfunction but clear consciousness.

Which time horizons of mortality are predicted?

Mortality is predicted for short-term outcomes at 3 and 7 days and relatively longer-term outcomes at 30 and 90 days after ICU admission.

How were predictors selected and how was performance evaluated?

The study evaluated 47 candidate variables and used the AutoScore framework to identify the top six predictors for each mortality model. Performance was assessed using the area under the ROC curve (AUC).

Aug 5, 2026

Upload by Rowan