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 Experiment Videos

Anatomically-guided masked autoencoder pre-training for aneurysm detection.

Alberto M Ceballos Arroyo1, Jisoo Kim2,3, Chu-Hsuan Lin2,3

  • 1Northeastern University.

IEEE Winter Conference on Applications of Computer Vision. IEEE Winter Conference on Applications of Computer Vision
|July 13, 2026
PubMed
Summary

This study introduces a new method for detecting intracranial aneurysms using AI. By pre-training a 3D Vision Transformer on head CT scans, it improves aneurysm detection sensitivity.

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

Biomechanical Comparison of Lumbar, Sacral and Iliac Screw Strain with the Novel Eiffel Tower Configuration Versus Conventional Pelvic Fixation in Adult Spinal Deformity Surgery.

Journal of surgical orthopaedic advances·2026
Same author

Publisher Correction: Tumor transcriptional state predicts survival in immune-checkpoint-blockade-treated glioblastoma.

Nature cancer·2026
Same author

A Bayesian approach to estimate ambient dose equivalent for liquid radioactive waste measurements.

Applied radiation and isotopes : including data, instrumentation and methods for use in agriculture, industry and medicine·2026
Same author

Tumor transcriptional state predicts survival in immune-checkpoint-blockade-treated glioblastoma.

Nature cancer·2026
Same author

Dietary intake patterns and nutritional adequacy in older adults with predialysis chronic kidney disease: a comparison by diabetes status.

Clinical nutrition research·2026
Same author

Homozygous CHD8 mutation intensifies ASD phenotypes and attenuates sex differences.

Molecular psychiatry·2026

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Radiology

Background:

  • Intracranial aneurysms pose significant global health risks, causing substantial morbidity and mortality.
  • Manual detection of intracranial aneurysms is challenging, time-consuming, and prone to error.
  • Limited annotated data hinders the development of effective automated aneurysm detection systems using traditional supervised learning.

Purpose of the Study:

  • To develop an automated method for detecting intracranial aneurysms.
  • To overcome the limitations of data scarcity in supervised learning for medical image analysis.
  • To improve the sensitivity and efficiency of aneurysm detection through novel pre-training strategies.

Main Methods:

  • Proposed a novel pre-training strategy for a 3D Vision Transformer model using unannotated head CT scans.

Related Experiment Videos

  • Modified masked auto-encoder (MAE) pre-training with factorized self-attention for computational viability in 3D.
  • Focused masked patches on arterial regions and incorporated artery distance maps for enhanced representation learning.

    • Main Results:

    • The proposed pre-training strategy significantly enhanced the performance of aneurysm detection models.
    • Achieved an absolute sensitivity gain of +2-7% compared to state-of-the-art (SOTA) models at a false positive rate of 0.5.
    • Demonstrated the effectiveness of reconstructing CT intensity values and artery distance maps.

    Conclusions:

    • The novel MAE-based pre-training approach effectively addresses data limitations in automated aneurysm detection.
    • The method offers a promising solution for improving the accuracy and efficiency of identifying intracranial aneurysms.
    • The developed technique has the potential to aid radiologists in faster and more reliable aneurysm diagnosis.