Previously submitted to: JMIR Formative Research (no longer under consideration since Nov 26, 2024)
Date Submitted: Jun 6, 2024
Warning: This is an author submission that is not peer-reviewed or edited. Preprints - unless they show as "accepted" - should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.
A Bayesian network approach for prognostic assessment in acute mild to moderate ischaemic stroke: Insights from a Multicentre Study
ABSTRACT
Background:
Mild to moderate ischaemic stroke accounts for one-half to two-thirds of all ischaemic stroke hospitalisations worldwide, yet prognostic models to assess functional outcome after stroke remain elusive.
Objective:
This study aimed to identify the risk factors and develop a prognostic assessment model for short-term functional outcome in patients with acute mild-to-moderate ischaemic stroke using the Bayesian Network (BN) approach.
Methods:
Patients participating in a multicentre randomised controlled trial were eligible for inclusion in this study. The least absolute shrinkage and selection operator (LASSO) regression was applied to reduce the dimensionality of the data, and synthetic minority oversampling technology (SMOTE) was used to correct data imbalances. The Bayesian network model was trained using a hill-climbing algorithm to determine the optimal network structure. The area under the receiver operating characteristic curve (AUROC) and the calibration plot were used to assess model performance.
Results:
A total of 2,432 patients were included for analysis, 424（17.43%）patients had unfavorable outcome in 3-month after AIS,. The BN model identified the NIH Stroke Scale (NIHSS) score, age, C-reactive protein (CRP) levels, fasting plasma glucose (FPG), and history of stroke as significant prognostic factors. The BN model outperformed the logistic regression model in accuracy, sensitivity, and F1 score in test sets, with significant differences in the AUROC as per the DeLong test (P < 0.05). The calibration plot showed good agreement between predictions and observations in both the training and test sets.
Conclusions:
The NIHSS score, age, CRP levels, FPG, and history of stroke are significant predictors of functional outcome in AIS patients. The BN model demonstrates superior predictive performance compared to traditional logistic regression, suggesting its potential as a valuable tool for assessing short-term prognosis in mild-moderate ischaemic stroke patients. Clinical Trial: The trial is registered in the Chinese Clinical Trial Registry (http://www. chictr.org.cn/, ChiCTR1900026492) on 12 October 2019.
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.