Related Experiment Video
Updated: Jan 6, 2026
09:10
Determining the Functional Status of the Corticospinal Tract Within One Week of Stroke
Published on: February 22, 2020
9.1K
StrokeENDPredictor-19: Setting New Prediction Model in Neurological Prognosis in Acute Ischemic Stroke.
Lingli Li1, Hongxiao Li2, Miaowen Jiang3
1Faculty of Information Science and Technology, Beijing University of Technology, Beijing, 100124, China.
Annals of Biomedical Engineering
|September 23, 2025
Summary
This study developed an AI model, StrokeENDPredictor-19, to accurately predict early neurological deterioration in acute ischemic stroke patients receiving IVT, aiding clinical decisions.
Area of Science:
- Neurology
- Artificial Intelligence
- Medical Informatics
Background:
- Early Neurological Deterioration (END) after intravenous thrombolysis (IVT) for acute ischemic stroke poses risks.
- Accurate prediction of END is crucial for optimizing treatment strategies.
Purpose of the Study:
- To develop an interpretable AI model for accurate prediction of END in acute ischemic stroke patients undergoing IVT.
- To enhance clinical decision-making by identifying patients at high risk of END.
Main Methods:
- A prospective cohort study of 970 acute ischemic stroke patients treated with IVT.
- Development and validation of five machine learning models, including XGBoost, using SHAP for interpretability.
- Internal and external validation of the StrokeENDPredictor-19 model.
Main Results:
- The XGBoost-based StrokeENDPredictor-19 model achieved high accuracy: 91% internally (AUC=0.96) and 90% externally (AUC=0.95).
- Established cutoff values for critical clinical features, offering practical reference standards.
- The model demonstrated superior performance compared to other developed models.
Conclusions:
- StrokeENDPredictor-19 provides a valuable tool for neurologists to forecast END risk in IVT-treated patients.
- The model supports more precise clinical decision-making for acute ischemic stroke management.
