Stroke Outcome Prediction using Vascular Imaging Features

Case ID:

This is a prediction method to determine the outcome of a stroke based on vascular images presented by the patient. This model predicts the 90-day functional outcome after ischemic stroke at a significantly higher accuracy than current models, to inform treatment decisions and patient selection. Machine learning models were employed and trained with patient-specific vascular imaging features, extracted using a cerebrovascular segmentation algorithm. Patient demographics and clinical history are also used as additional predictors.

Stroke is the leading cause of death in the U.S. currently. Not only has it cost several thousands of lives but also billions of dollars are spent on treating patients who suffered from a stroke and most of these patients are unaware of the major symptoms, which increases their chance of disability. Researchers from all over the country have been conducting studies on stroke prevention in various methods. One of these methods is predicting the outcome through an app or software. Many studies have shown an emphasis on the importance of vascular imaging in predicting stroke outcomes. There are also similar products in the market that provide stroke prediction using imaging features and patient demographics. 


  • Stroke prevention


  • Higher accuracy 
  • Help make better treatment decisions
Patent Information:
Contact For More Information:
Tariq Ahmed
Sr Licensing Manager, College of Engineering
The University of Arizona
Lead Inventor(s):
Kaveh Laksari
Aditi Deshpande
Jordan Elliott
Pouya Tahsili-Fahadan