Development and Validation of an Interpretable Machine Learning Model for Prediction of the Risk of Clinically Ineffective Reperfusion in Patients Following Thrombectomy for Ischemic Stroke

Aug 2, 2026

Upload by Miles

Despite successful recanalization after thrombectomy in acute ischemic stroke, clinically ineffective reperfusion (CIR) still leaves patients with poor outcomes. This study assembled thrombectomy patient data from December 2021 to June 2024, compared clinical variables between CIR and effective reperfusion groups, and trained four machine-learning classifiers. Random forest achieved the strongest test performance (AUC 0.96). SHAP-based interpretation highlighted endovascular thrombectomy attempts as the key driver, and a web-based calculator enables early, personalized risk estimation and intervention planning.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages11

Reading Time28 min

AccessFree to download

Background

Methods

Data collection and cohort

Model development and evaluation

Results

Model performance comparison

Feature importance and key determinants

Conclusion

Keywords

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 an Interpretable Machine Learning Model for Prediction of the Risk of Clinically Ineffective Reperfusion in Patients Following Thrombectomy for Ischemic Stroke Xiaolong Hu 1 , 2 , *, Dayong Qi2 , *, Suya Li 2 , Shifei Ye2 , Yue Chen 2 , Wei Cao2 , Meng Du2 , Tianheng Zheng2 , Peng Li 2 , Yibin Fang 1–3 1Tongji University School of Medicine, Tongji University Affiliated Shanghai 4th People’s Hospital, Shanghai, People’s Republic of China; 2Department of Neurovascular Disease, Tongji University Affiliated Shanghai 4th People’s Hospital, Shanghai, People’s Republic of China; 3Translational Research Institute of Brain and Brain-Like Intelligence, Shanghai Fourth People’s Hospital, School of Medicine, Tongji University, Shanghai, People’s Republic of China *These authors contributed equally to this work Correspondence: Yibin Fang, Tongji University Affiliated Shanghai 4th People’s Hospital, Sanmen Road 1279, Hongkou District, Shanghai, People’s Republic of China, Email [fangyibin4@163.com](fangyibin4@163.com) | Background: Despite successful recanalization after thrombectomy in patients with acute ischemic stroke, poor prognosis often persists. This study aimed to investigate the factors contributing to clinically ineffective reperfusion (CIR), develop and validate a machine-learning model to predict CIR, and provide guidance for future clinical treatments.
Methods: We collected data from patients undergoing thrombectomy at Shanghai Fourth People’s Hospital between December 2021 and June 2024. The clinical variables were compared between the clinically ineffective and effective recanalization groups using univariate analysis. Four machine learning models were developed: random forest (RF), support vector machine (SVM), decision tree (DT), and k-nearest neighbor (KNN) . Model performance was evaluated using receiver operating characteristic (ROC) curves and heatmap visualization. The SHAP method rank the feature importance and provided interpretability for the final model.
Results: Among the four machine learning models, the RF model showed the best performance, with an area under the curve (AUC) of 0.96 (95% CI: 0.91–1.0), accuracy of 0.93, and specificity of 0.97 on the test dataset. The SHAP algorithm identified the number of endovascular thrombectomy (EVT) attempts as the key factor influencing CIR. Based on the RF model, a web-based calculator for CIR prediction is available at [https://ineffectivereperfusion.shinyapps.io/calculate/](https://ineffectivereperfusion.shinyapps.io/calculate/. The final model included ten parameters: EVT)[. The final model included ten parameters: EVT](https://ineffectivereperfusion.shinyapps.io/calculate/. The final model included ten parameters: EVT)[ ](https://ineffectivereperfusion.shinyapps.io/calculate/. The final model included ten parameters: EVT)attempts, diabetes mellitus, previous ischemic stroke, National Institutes of Health Stroke Scale (NIHSS score), preoperative infarction in the basal ganglia, baseline diastolic blood pressure, clot burden score (CBS)/basilar artery on computed tomography angiography (BATMAN) score, stroke cause, collateral grade, and MLS.
Conclusion: We developed and validated the first interpretable machine learning model for CIR prediction after EVT, surpassing traditional methods. Our CIR risk prediction platform enables early intervention and personalized treatment. The number of EVT attempts has emerged as a key determinant, underscoring the need for optimized procedural timing to improve outcomes. Keywords: machine learning, clinically ineffective reperfusion, predictive model, acute ischemic stroke, online predictive platform | | --- | | Introduction
Acute ischemic stroke

FAQs

What was the goal of developing the interpretable machine learning model?

To identify factors associated with clinically ineffective reperfusion and to build and validate a model that predicts CIR risk after thrombectomy for ischemic stroke.

Which machine learning approach performed best and how was it evaluated?

The random forest model showed the best performance, evaluated using ROC curves, accuracy, specificity, and AUC on the test dataset, achieving AUC 0.96.

How does the study interpret the model’s predictions?

The SHAP method ranks feature importance to provide interpretability for the final model, clarifying which variables most influence CIR.

What is the key predictor identified for clinically ineffective reperfusion?

The number of endovascular thrombectomy (EVT) attempts was identified as the key factor affecting CIR risk.

Aug 2, 2026

Upload by Miles