Related Experiment Video
Updated: Sep 7, 2025
Permanent Cerebral Vessel Occlusion via Double Ligature and Transection
Published on: July 21, 2013
Preprocedural determination of an occlusion pathomechanism in endovascular treatment of acute stroke: a machine
Jang-Hyun Baek1,2, Byung Moon Kim3, Dong Joon Kim4
1Department of Neurology, Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea.
Objective:
To evaluate whether an occlusion pathomechanism can be accurately determined by common preprocedural findings through a machine learning-based prediction model (ML-PM).
Methods:
A total of 476 patients with acute stroke who underwent endovascular treatment were retrospectively included to derive an ML-PM. For external validation, 152 patients from another tertiary stroke center were additionally included. An ML algorithm was trained to classify an occlusion pathomechanism into embolic or intracranial atherosclerosis. Various common preprocedural findings were entered into the model. Model performance was evaluated based on accuracy and area under the receiver operating characteristic curve (AUC). For practical utility, a decision flowchart was devised from an ML-PM with a few key preprocedural findings. Accuracy of the decision flowchart was validated internally and externally.
Results:
An ML-PM could determine an occlusion pathomechanism with an accuracy of 96.9% (AUC=0.95). In the model, CT angiography-determined occlusion type, atrial fibrillation, hyperdense artery sign, and occlusion location were top-ranked contributors. With these four findings only, an ML-PM had an accuracy of 93.8% (AUC=0.92). With a decision flowchart, an occlusion pathomechanism could be determined with an accuracy of 91.2% for the study cohort and 94.7% for the external validation cohort. The decision flowchart was more accurate than single preprocedural findings for determining an occlusion pathomechanism.
Conclusions:
An ML-PM could accurately determine an occlusion pathomechanism with common preprocedural findings. A decision flowchart consisting of the four most influential findings was clinically applicable and superior to single common preprocedural findings for determining an occlusion pathomechanism.
Insights
Machine learning accurately predicts stroke occlusion causes using preprocedural findings. A decision flowchart based on key factors improves clinical applicability for determining embolic versus intracranial atherosclerosis mechanisms.
Area of Science:
- Neurology
- Medical Imaging
- Artificial Intelligence
Background:
- Determining the occlusion pathomechanism in acute stroke is crucial for guiding endovascular treatment.
- Common preprocedural findings are often used, but their predictive accuracy for specific pathomechanisms can vary.
Purpose of the Study:
- To develop and validate a machine learning-based prediction model (ML-PM) to accurately determine stroke occlusion pathomechanisms.
- To assess the utility of common preprocedural findings in predicting whether a stroke is embolic or due to intracranial atherosclerosis.
Main Methods:
- A machine learning algorithm was trained on data from 476 acute stroke patients undergoing endovascular treatment.
- The model classified occlusion pathomechanisms (embolic vs. intracranial atherosclerosis) using various preprocedural findings.
- External validation was performed on 152 additional patients; a decision flowchart was created for clinical application.
Main Results:
- The ML-PM achieved high accuracy (96.9%) and AUC (0.95) in determining occlusion pathomechanisms.
- Key predictors included CT angiography occlusion type, atrial fibrillation, hyperdense artery sign, and occlusion location.
- A decision flowchart using these four factors showed strong accuracy (91.2% internal, 94.7% external validation) and outperformed individual findings.
Conclusions:
- Machine learning models can accurately predict stroke occlusion pathomechanisms using readily available preprocedural data.
- A simplified decision flowchart derived from the ML-PM is clinically applicable and enhances diagnostic accuracy compared to relying on single findings.
