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

In Vivo Wound Healing Activity of <i>Abrus cantoniensis</i> Extract.

Evidence-based complementary and alternative medicine : eCAM·2017
Same author

Gigantol from Dendrobium chrysotoxum Lindl. binds and inhibits aldose reductase gene to exert its anti-cataract activity: An in vitro mechanistic study.

Journal of ethnopharmacology·2017
Same author

[High-risk factors and clinical characteristics of massive pulmonary hemorrhage in infants with extremely low birth weight].

Zhongguo dang dai er ke za zhi = Chinese journal of contemporary pediatrics·2017
Same author

Liver X receptor agonist T0901317 reverses resistance of A549 human lung cancer cells to EGFR-TKI treatment.

FEBS open bio·2017
Same author

Naked eye plasmonic indicator with multi-responsive polymer brush as signal transducer and amplifier.

Nanoscale·2017
Same author

Graphene oxide adsorbent based dispersive solid phase extraction coupled with multi-pretreatment clean-up for analysis of trace aflatoxins in traditional proprietary Chinese medicines.

Journal of chromatography. B, Analytical technologies in the biomedical and life sciences·2017

Related Experiment Video

Updated: Mar 31, 2026

A Volumetric Method for Quantification of Cerebral Vasospasm in a Murine Model of Subarachnoid Hemorrhage
08:12

A Volumetric Method for Quantification of Cerebral Vasospasm in a Murine Model of Subarachnoid Hemorrhage

Published on: July 28, 2018

8.5K

Cerebrovascular diagnosis using CTA-based intracranial aneurysm classification via transfer learning and Grad-CAM

Xin Wang1, Dan Chen2, Kavimbi Chipusu3

  • 1Imaging Center, The Third People's Hospital of Hefei, Hefei Third Clinical College of Anhui Medical University, Hefei, China.

Frontiers in Neurology
|March 30, 2026
PubMed
Summary

A new deep transfer learning framework improves intracranial aneurysm classification from CT angiography scans. This AI approach enhances accuracy and provides interpretable results, aiding neurovascular imaging diagnostics.

Keywords:
Grad-CAM visualizationcomputed tomography angiographydeep transferlearningintracranial aneurysm classificationneuro-vascular bioengineering

More Related Videos

Comprehensive Endovascular and Open Surgical Management of Cerebral Arteriovenous Malformations
14:58

Comprehensive Endovascular and Open Surgical Management of Cerebral Arteriovenous Malformations

Published on: October 20, 2017

10.4K
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

12.8K

Related Experiment Videos

Last Updated: Mar 31, 2026

A Volumetric Method for Quantification of Cerebral Vasospasm in a Murine Model of Subarachnoid Hemorrhage
08:12

A Volumetric Method for Quantification of Cerebral Vasospasm in a Murine Model of Subarachnoid Hemorrhage

Published on: July 28, 2018

8.5K
Comprehensive Endovascular and Open Surgical Management of Cerebral Arteriovenous Malformations
14:58

Comprehensive Endovascular and Open Surgical Management of Cerebral Arteriovenous Malformations

Published on: October 20, 2017

10.4K
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

12.8K

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Neurosurgery

Background:

  • Intracranial aneurysms (IA) affect 2-5% of the population, posing significant mortality risks upon rupture.
  • Accurate classification of IA from computed tomography angiography (CTA) is vital for patient management but faces challenges with small datasets and limited interpretability.
  • Deep learning models offer potential for IA classification but require enhancement for clinical utility.

Purpose of the Study:

  • To evaluate a hybrid deep transfer learning framework integrated with Grad-CAM for improved discrimination and explainability in CTA-based IA classification.
  • To enhance the accuracy and transparency of AI-driven IA classification, particularly in data-limited scenarios.
  • To provide a validated, interpretable AI tool for neurovascular imaging analysis.

Main Methods:

  • A retrospective study involving 83 patients and CTA data from two centers.
  • Comparison of a baseline deep learning (DL) model, a transfer learning-enhanced model (DL+TL), and radiologist assessment using stratified 5-fold cross-validation.
  • AI models utilized a hybrid ResNet-18 architecture with LASSO feature selection and logistic regression; interpretability assessed via Grad-CAM (IoU, Dice score).

Main Results:

  • The DL+TL model achieved superior performance with a mean AUC of 0.853 and accuracy of 84.0%, outperforming DL (AUC: 0.744) and radiologists (AUC: 0.731).
  • Grad-CAM analysis demonstrated significantly higher attention precision for DL+TL (IoU: 0.68) compared to DL (IoU: 0.45).
  • Blinded radiologists rated the DL+TL model's interpretability as more clinically relevant (4.2/5) than the DL model (2.8/5).

Conclusions:

  • Integrating transfer learning with quantitative interpretability assessment enhances both accuracy and transparency in IA classification.
  • The developed framework offers a validated, interpretable approach for neurovascular imaging.
  • Further multi-center validation is recommended to confirm the clinical applicability of this AI framework.