Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Intensive Versus Conventional Blood Pressure Lowering After Successful Endovascular Thrombectomy: OPTIMAL-BP 1-Year Outcomes.

Stroke·2026
Same author

Prognostic Modeling Based on Post-Endovascular Thrombectomy Systolic Blood Pressure Trajectories Using Explainable Artificial Intelligence: A Secondary Analysis of the OPTIMAL-BP Trial.

Journal of medical systems·2026
Same author

Poorer Outcomes With Blood Pressure Lowering in Severe White Matter Disease After Thrombectomy.

Stroke·2026
Same author

Multimodal Antiatherosclerotic Effects of Clinical-Grade Mesenchymal Stem Cell-Derived Extracellular Vesicles.

Stroke·2026
Same author

Supracardiac Atherosclerotic Lesions in Embolic Stroke of Undetermined Source: Predicting Stroke Recurrence and Enhancing Secondary Prevention Strategies.

Journal of the American Heart Association·2025
Same author

Nano-Encapsulated Spicule System Enhances Delivery of Wharton's Jelly MSC Secretome and Promotes Skin Rejuvenation: Preclinical and Clinical Evaluation.

International journal of molecular sciences·2025

Related Experiment Video

Updated: Sep 22, 2025

A Fibrin-Enriched and tPA-Sensitive Photothrombotic Stroke Model
09:42

A Fibrin-Enriched and tPA-Sensitive Photothrombotic Stroke Model

Published on: June 4, 2021

2.9K

A Deep Learning-Based Automatic Collateral Assessment in Patients with Acute Ischemic Stroke.

Yoon-Chul Kim1, Jong-Won Chung2, Oh Young Bang3

  • 1Division of Digital Healthcare, College of Software and Digital Healthcare Convergence, Yonsei University, Wonju, Republic of Korea.

Translational Stroke Research
|May 21, 2022
PubMed
Summary

A deep learning model accurately grades collateral status in acute ischemic stroke patients using perfusion imaging. This AI tool shows potential to assist in assessing stroke severity and guiding treatment decisions.

Keywords:
Deep learningIschemiaMagnetic resonance imagingPerfusionStroke

More Related Videos

Optimized Management of Endovascular Treatment for Acute Ischemic Stroke
09:21

Optimized Management of Endovascular Treatment for Acute Ischemic Stroke

Published on: January 18, 2018

12.2K
Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
10:25

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

Published on: September 25, 2019

48.4K

Related Experiment Videos

Last Updated: Sep 22, 2025

A Fibrin-Enriched and tPA-Sensitive Photothrombotic Stroke Model
09:42

A Fibrin-Enriched and tPA-Sensitive Photothrombotic Stroke Model

Published on: June 4, 2021

2.9K
Optimized Management of Endovascular Treatment for Acute Ischemic Stroke
09:21

Optimized Management of Endovascular Treatment for Acute Ischemic Stroke

Published on: January 18, 2018

12.2K
Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
10:25

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

Published on: September 25, 2019

48.4K

Area of Science:

  • Neurology
  • Radiology
  • Artificial Intelligence

Background:

  • Assessing collateral status in large vessel occlusion (LVO) acute ischemic stroke (AIS) is crucial for predicting outcomes.
  • Dynamic susceptibility contrast magnetic resonance perfusion (DSC-MRP) imaging provides valuable data for collateral assessment.
  • Manual grading of collateral status by experts can be time-consuming and subject to inter-observer variability.

Purpose of the Study:

  • To develop and validate a supervised deep learning (DL) model for automated collateral status grading from DSC-MRP images in LVO-AIS patients.
  • To compare the performance of the DL model against expert manual grading.
  • To evaluate the association of DL-graded collateral status with infarct growth and clinical outcomes.

Main Methods:

  • A supervised DL model was developed using DSC-MRP data from 255 LVO-AIS patients, trained to classify collateral status as good or poor based on expert readings.
  • The model was externally validated on a separate cohort of 72 LVO-AIS patients.
  • Performance was assessed using c-statistics and kappa statistics for agreement with expert grades.

Main Results:

  • The DL model achieved a c-statistic of 0.91 in internal validation and 0.85 in external validation.
  • Moderate agreement (kappa = 0.53) was observed between DL-based and expert grading in the external validation cohort.
  • Poor collateral status, as graded by the DL model, was associated with higher Day 7 infarct growth (26 mL vs. 6 mL) and less favorable 90-day functional outcomes (common odds ratio = 2.99).

Conclusions:

  • The developed DL model demonstrates robust performance in grading collateral status from DSC-MRP images in LVO-AIS patients.
  • The DL-based grading shows good agreement with expert manual assessment.
  • This AI tool has the potential to aid in the rapid and objective assessment of collateral status, potentially improving stroke management and patient outcomes.