Predicting Mechanical Thrombectomy Outcomes with Smart, Transparent AI
Official title: Probabilistic and Explainable Machine Learning Methods for the Prediction of Technical Success and Stroke Outcomes Following Mechanical Thrombectomy
Funded at Northwestern University · NIH-11290590 · Data updated 2026-11-12
A patient-friendly look at how new AI models may predict how well mechanical thrombectomy could work for individuals with acute ischemic stroke.
Quick facts
| Grant type | R21 grant |
|---|---|
| Study type | NIH-funded research |
| Funding institution | Northwestern University NIH-funded |
| Lab location | 1 site (Chicago, United States) |
| Project start | July 21, 2026 |
| Project end | May 31, 2028 |
| Funding year | 2026 |
| Award (this year) | $120,000 |
| NIH project number | 1R21HL184309-01 |
| Project ID | NIH-11290590 on NIH RePORTER |
Who could benefit from this research
✓Could help
Adults with acute ischemic stroke due to large vessel occlusion and considered for mechanical thrombectomy.
✕Less likely to help
Patients who are not candidates for thrombectomy or whose clinical condition is rapidly changing may not benefit from this predictive approach.
What this research studies
This project develops probabilistic, explainable machine learning tools to predict how many retrieval attempts and what outcomes a patient might have after a clot-removal procedure for acute stroke. The goal is to give clinicians and patients a clearer sense of the likely range of results for an individual, not just an average. Researchers will train models on clinical data and clot/anatomy features to produce outcome distributions that are easy to understand and trust. The work emphasizes transparency, so the models explain why they make certain predictions and how uncertainty is handled.
Why it matters
Potential benefit: If successful, the approach could help doctors tailor decisions and set appropriate expectations for patients and families after mechanical thrombectomy.
How similar studies have gone: Earlier work used traditional models to predict outcomes; this project adds probabilistic and explainable elements that may offer more nuanced guidance.
Where this research is happening
-
Northwestern University
Chicago, United States Active
Researchers
- Principal investigator: Cantrell, Donald Robinson — Northwestern University
- Study coordinator: Cantrell, Donald Robinson
Related research
Other NIH-funded projects on the same topics.
-
Robotic ultrasound-guided clot removal for stroke treatment
A new robot-assisted ultrasound system is being developed to help quickly reopen brain arteries during a stroke, with testing in patient-specific 3D-printed models and animal studies.
-
Modeling Acute Ischemic Stroke to Improve Mechanical Thrombectomy
A research project is creating better computer models and animal studies to help remove clots more safely and effectively in people who have an acute ischemic stroke.
-
Predictors of recovery after stroke in rural Appalachia
This project tests blood protein markers to predict cognitive recovery after a stroke treated with thrombectomy, focusing on rural Appalachian patients.
-
Improving stroke treatment with a new catheter device for easier access.
This study is testing a new catheter device that helps doctors remove blood clots in the brain through the wrist instead of the groin, aiming to make the treatment for stroke patients safer and more effective.
-
Improving treatment for acute ischemic stroke using advanced modeling techniques
This study is looking at how blood clots form in the brain and why some can be removed easily while others can't, with the goal of helping doctors improve treatments for people who have had a stroke.
-
Developing a smart catheter to improve blood clot removal in stroke treatment
This study is testing a new smart catheter that helps doctors remove blood clots during stroke treatment by giving them real-time information about how the procedure is going, making it safer and more effective for patients.
About this research
- This is an active NIH-funded research project — typically early-stage science, not a clinical trial accepting patient enrollment.
- Some NIH-funded labs run parallel clinical studies or seek volunteers for related work. To check, contact the principal investigator or institution listed above.
- For full project details, budget, and progress reports, visit the official NIH RePORTER page.