Related Experiment Video
Updated: May 27, 2025
09:52
Setting Up a Stroke Team Algorithm and Conducting Simulation-based Training in the Emergency Department - A Practical Guide
Published on: January 15, 2017
17.1K
World's First Real-Time Artificial Intelligence-Assisted Mechanical Thrombectomy for Acute Ischemic Stroke
Eisuke Hirose1, Yoshikazu Matsuda2, Syun Takano1
1From the Department of Neurosurgery (E.H., Y.M., S.T., R.A., T.M.), Showa Medical University Hospital, Tokyo, Japan.
AJNR. American Journal of Neuroradiology
|February 17, 2025
Summary
This study shows real-time artificial intelligence (AI) assistance is accurate and safe for mechanical thrombectomy (MT) in acute ischemic stroke (AIS). The AI system effectively alerted clinicians to guiding catheter (GC) repositioning needs during emergent procedures.
Area of Science:
- Neurosurgery
- Medical Imaging
- Artificial Intelligence
Background:
- Real-time artificial intelligence (AI) is emerging as a tool in neuroendovascular procedures.
- Previous applications include carotid artery stenting and aneurysm embolization.
Purpose of the Study:
- To evaluate the preliminary experience and effectiveness of real-time AI-assisted mechanical thrombectomy (MT) for acute ischemic stroke (AIS).
- To assess the accuracy and safety of AI-driven notifications for guiding catheter (GC) repositioning during emergent MT procedures.
Main Methods:
- A deep learning-based AI software (Neuro-Vascular Assist) was used during MT for 16 consecutive AIS patients.
- The AI system provided real-time alerts for GC movement outside the fluoroscopic view.
- Efficacy, accuracy, and safety were evaluated using video recordings and procedural data.
Main Results:
- The AI system functioned properly in all cases, with a mean of 8.1 notifications per case for GC repositioning.
- AI notification precision was 97% and recall was 99%.
- 20% of true-positive alerts prompted timely GC repositioning, indicating potential clinical utility without adverse events or delays.
Conclusions:
- Real-time AI software demonstrated sufficient accuracy and safety for emergent MT in AIS.
- The AI system shows promise for improving procedural workflow and patient outcomes in acute stroke interventions.
- Further large-scale studies are needed to validate the impact of AI assistance in MT procedures.
