ORIGINAL RESEARCH article
Front. Neurol.
Sec. Neurorehabilitation
Construction and Validation of a Prognostic Model for Clinical Outcomes in Patients with Prolonged Disorders of Consciousness Based on Multidimensional Indicators: A Prospective Cohort Study
Zhen Feng
Qiaojun Zhang
Second Affiliated Hospital of Xi'an Jiaotong University, Xi'an, China
Select one of your emails
You have multiple emails registered with Frontiers:
Notify me on publication
Please enter your email address:
If you already have an account, please login
You don't have a Frontiers account ? You can register here
Abstract
The prognostic assessment of patients with prolonged Disorders of Consciousness (pDoC) remains challenging due to the subjectivity, high costs, and limitations of existing methods. This study aims to develop and validate a prognostic model for pDoC patients using multidimensional clinical indicators to improve accuracy, objectivity, and clinical decision-making. A prospective cohort study involving 304 pDoC patients from the First Affiliated Hospital of Nanchang University was conducted between January 2021 and October 2023. Data collected included demographics, etiology, disease course, behavioral scales (GCS, FOUR, CRS-R), laboratory indicators, and other clinical information. Prognosis was categorized into good (GOS-E 3-8) and poor (GOS-E 1-2) outcomes using the Glasgow Outcome Scale-Extended (GOS-E). Key predictive factors were identified through the Boruta algorithm and recursive feature elimination (RFE). Machine learning models (logistic regression, SVM, MLP, XGBoost, GBM) were developed with scikit-learn, and performance was evaluated by accuracy, ROC-AUC, and decision curve analysis (DCA). SHAP analysis was employed for model interpretability. The Gradient Boosting Machine (GBM) model outperformed others, achieving the highest AUROC of 0.954 (95% CI: 0.924-0.977) in the training set and 0.922 (95% CI: 0.847-0.979) in the test set. DCA confirmed significant net benefit across most thresholds. CRS-R score, age, FOUR score, GCS total, and hospitalization length were identified as key prognostic factors. In conclusion, a robust prognostic model for pDoC patients using multidimensional clinical data and machine learning has been developed and validated. The GBM model demonstrated excellent performance, providing an objective, accurate tool for prognosis and guiding individualized treatment and rehabilitation. Further multi-center studies are needed to optimize and validate this model.
Summary
Keywords
Behavioral scales, clinical evaluation, disorders of consciousness, Prediction model, prognosis
Received
03 November 2025
Accepted
30 January 2026
Copyright
© 2026 Feng and Zhang. 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) or licensor 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.
*Correspondence: Qiaojun Zhang
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.