Development and Validation of Machine Learning Models for Outcome Prediction in Patients with Poor-Grade Aneurysmal Subarachnoid Hemorrhage Following Endovascular Treatment - ORIGINAL RESEARCH

Aug 2, 2026

Upload by Sage

Endovascular treatment (EVT) is recommended for intracranial aneurysms, yet poor-grade aneurysmal subarachnoid hemorrhage (aSAH) still shows a high rate of unfavorable functional outcome. This study extracts outcome-related variables from the multicenter PROSAH-MPC registry, splits data into training and validation cohorts (7:3), and builds nine machine learning models plus a stack ensemble. Multivariable logistic regression identifies independent risk factors, and model performance is assessed with AUC-ROC, complemented by SHAP-based feature explanation. LightGBM demonstrates the best validation performance and highlights age as a key driver.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages15

Reading Time38 min

AccessFree to download

Introduction

Background and rationale

Methods

Dataset and cohort assignment

Model development and evaluation

Explainability using SHAP

Results

Cohort outcomes

Independent risk factors

Predictive performance and feature importance

Conclusion

Therapeutics and Clinical Risk Management downloaded from [https://www.dovepress.com/](https://www.dovepress.com/) For personal use only. Therapeutics and Clinical Risk Management Open Access Full Text Article ORIGINAL RESEARCH Development and Validation of Machine Learning Models for Outcome Prediction in Patients with Poor-Grade Aneurysmal Subarachnoid Hemorrhage Following Endovascular Treatment Senlin Du 1–4 , *, Yanze Wu 1–4 , *, Jiarong Tao 1 , Lei Shu 1–4 , Tengfeng Yan 1–4 , Bing Xiao 1 , Shigang Lv 1 , Minhua Ye 1 , Yanyan Gong 1 , Xingen Zhu 1–4 , Ping Hu 1 , 5 , Miaojing Wu 1 1Department of Neurosurgery, The second Affiliated Hospital, Jiangxi Medical College of Nanchang University, Nanchang, 330006, People’s Republic of China; 2Jiangxi Key Laboratory of Neurological Tumors and Cerebrovascular Diseases, Nanchang, 330006, People’s Republic of China; 3Jiangxi Health Commission Key Laboratory of Neurological Medicine, Nanchang, 330006, People’s Republic of China; 4Institute of Neuroscience, Nanchang University, Nanchang, 330006, People’s Republic of China; 5Department of Neurosurgery, Panzhihua Central Hospital, The second Clinical Medical College of Panzhihua University, Panzhihua, 617067, People’s Republic of China *These authors contributed equally to this work Correspondence: Miaojing Wu, Department of Neurosurgery, The second Affiliated Hospital, Jiangxi Medical College of Nanchang University, Nanchang, 330006, People’s Republic of China, Email wmj [1987@163.com](1987@163.com) | Background: Endovascular treatment (EVT) has been recommended as a superior modality for the treatment of intracranial aneurysm. However, there still exists a worse percentage of poor functional outcome in patients with poor-grade aneurysmal subarachnoid hemorrhage (aSAH) undergoing EVT. Therefore, it is urgently needed to investigate the risk factors and develop a critical decision model in the subtype of such patients.
Methods: We extracted the target variables from an ongoing registry cohort study, PROSAH-MPC, which was conducted in multiple centers in China. We randomly assigned these patients to training and validation cohorts with a ratio of 7:3 . Univariate and multivariate logistic regressions were performed to find the potential factors, and then nine machine learning models and a stack ensemble model were developed with optimized variables. The performance of these models was evaluated through several indicators, including area under the receiver operating characteristic curve (AUC-ROC) . We further use Shapley Additive Explanations (SHAP) methods for the distribution of feature visualization based on the optimal models.
Results: A total of 226 eligible patients with poor-grade aSAH undergoing EVT were enrolled, while 89 (39.4%) has a poor 12-month outcome. Age (Adjusted OR [aOR], 1.08; 95% CI: 1.03–1.13; p = 0.002), subarachnoid hemorrhage volume (aOR, 1.02; 95% CI: 1.00–1.05; p = 0.033), World Federation of Neurosurgical Societies grade (WFNS) (aOR, 2.03; 95% CI: 1.05–3.93; p = 0.035), and Hunt-Hess grade (aOR, 2.36; 95% CI: 1.13–4.93; p = 0.022) were identified as the independent risk factors of the poor outcome. Then, the prediction models developed have revealed that LightGBM algorithm has a superior performance with an AUC-ROC value of 0.842 in the validation cohort, while the SHAP results showed that age is the most important risk factor affecting functional outcomes. Conclusion: The LightGBM model holds immense potential in facilitating risk stratification for poor-grade aSAH patients undergoing endovascular treatment who are at risk of adverse outcomes, thereby enhancing clinical decision-making processes.
Trial Registration: PROSAH-MPC. NCT05738083. Registered 16 November 2022 – Retrospectively registered, [https://clinical](https://clinical)[ ](https://clinical)[trials.gov/study/NCT05738083](trials.gov/study/NCT05738083) .
Keywords: intracranial aneurysm, subarachnoid hemorrhage, endovascu

FAQs

What clinical problem does the study address in poor-grade aSAH patients?

It targets the persistently high proportion of poor functional outcomes in patients with poor-grade aneurysmal subarachnoid hemorrhage undergoing endovascular treatment.

How were the machine learning models developed and validated?

Variables were extracted from the multicenter PROSAH-MPC registry, patients were randomly split into training and validation cohorts in a 7:3 ratio, and logistic regression was used to identify factors before training nine machine learning models and a stack ensemble.

Which model performed best and which factor was most influential?

LightGBM achieved superior validation performance with an AUC-ROC of 0.842, and SHAP results indicated that age is the most important risk factor affecting functional outcomes.

Aug 2, 2026

Upload by Sage