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

Computed Tomography01:10

Computed Tomography

4.2K
Tomography refers to imaging by sections. Computed tomography (CT) is a non-invasive imaging technique that uses computers to analyze several cross-sectional X-rays to reveal minute details about structures in the body.
The technique was invented in the 1970s and is based on the principle that as X-rays pass through the body, they are absorbed or reflected at different levels. In the technique, a patient lies on a motorized platform while a computerized axial tomography (CAT) scanner rotates...
4.2K
Brain Imaging01:14

Brain Imaging

200
Brain imaging technologies provide critical insights into both the structure and function of the human brain, enabling medical professionals and researchers to diagnose, study, and treat neurological disorders or psychiatric disorders more effectively.
These technologies include computerized axial tomography (CAT or CT scans), positron-emission tomography (PET scans),  magnetic resonance imaging (MRI),  functional magnetic resonance imaging (fMRI), and Transcranial Magnetic...
200
Magnetic Resonance Imaging01:24

Magnetic Resonance Imaging

4.9K
Magnetic resonance imaging (MRI) is a noninvasive medical imaging technique based on a phenomenon of nuclear physics discovered in the 1930s, in which matter exposed to magnetic fields and radio waves was found to emit radio signals. In 1970, a physician and researcher named Raymond Damadian noticed that malignant (cancerous) tissue gave off different signals than normal body tissue. He applied for a patent for the first MRI scanning device in clinical use by the early 1980s. The early MRI...
4.9K

You might also read

Related Articles

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

Sort by
Same author

MAD-MT Score: A Tool to Optimize Patient Selection for Mechanical Thrombectomy in Distal Vessel Occlusions.

Stroke·2026
Same author

Sex Differences in the Protective Effect of Brain Volume: Age Attenuates Protection in Women.

Stroke (Hoboken, N.J.)·2026
Same author

Thrombectomy in Posterior Circulation Tandem Occlusions: Multicenter Comparative Analysis of Procedural Techniques and Predictors of Clinical Outcomes.

Radiology·2026
Same author

Context-aware multi-property antibody predictor: a novel framework integrating text and protein language models.

NPJ systems biology and applications·2026
Same author

Intra-Arterial Thrombolytic vs. Mechanical Thrombectomy in Distal Medium Middle Cerebral Artery Acute Ischemic Stroke: A Preliminary Multinational Multicenter Propensity Score-Matched Study.

Clinical neuroradiology·2026
Same author

Sex-Related Differences in Outcomes of Endovascular Treatment of Distal Medium Vessel Occlusion Strokes.

Clinical neuroradiology·2026

Related Experiment Video

Updated: May 23, 2025

3D Scanning Technology Bridging Microcircuits and Macroscale Brain Images in 3D Novel Embedding Overlapping Protocol
10:14

3D Scanning Technology Bridging Microcircuits and Macroscale Brain Images in 3D Novel Embedding Overlapping Protocol

Published on: May 12, 2019

7.2K

Encoding 3D information in 2D feature maps for brain CT-Angiography.

Uma M Lal-Trehan Estrada1, Sunil Sheth2, Arnau Oliver1

  • 1Research institute of Computer Vision and Robotics, University of Girona, Girona, Spain.

Computerized Medical Imaging and Graphics : the Official Journal of the Computerized Medical Imaging Society
|March 11, 2025
PubMed
Summary
This summary is machine-generated.

Learnable 3D pooling (L3P) efficiently compresses 3D brain scans into 2D maps for improved large vessel occlusion detection and brain age prediction. This method matches 3D model performance with fewer resources, enhancing interpretability.

Keywords:
2D feature maps3D-to-2DBrain imagingLarge vessel occlusionLearnable 3D pooling

More Related Videos

A Pipeline for 3D Multimodality Image Integration and Computer-assisted Planning in Epilepsy Surgery
09:41

A Pipeline for 3D Multimodality Image Integration and Computer-assisted Planning in Epilepsy Surgery

Published on: May 20, 2016

12.0K
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

47.7K

Related Experiment Videos

Last Updated: May 23, 2025

3D Scanning Technology Bridging Microcircuits and Macroscale Brain Images in 3D Novel Embedding Overlapping Protocol
10:14

3D Scanning Technology Bridging Microcircuits and Macroscale Brain Images in 3D Novel Embedding Overlapping Protocol

Published on: May 12, 2019

7.2K
A Pipeline for 3D Multimodality Image Integration and Computer-assisted Planning in Epilepsy Surgery
09:41

A Pipeline for 3D Multimodality Image Integration and Computer-assisted Planning in Epilepsy Surgery

Published on: May 20, 2016

12.0K
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

47.7K

Area of Science:

  • Medical Imaging
  • Computer Vision
  • Artificial Intelligence

Background:

  • Analyzing 3D neuroimaging data (CTA, MRI) for conditions like large vessel occlusion (LVO) and brain age prediction presents computational challenges.
  • Existing methods often require significant resources or struggle to effectively capture 3D spatial information in 2D representations.

Purpose of the Study:

  • To introduce Learnable 3D Pooling (L3P), a novel Convolutional Neural Network (CNN) module for compressing 3D volumetric data into 2D feature maps.
  • To evaluate L3P's effectiveness in improving predictions for 3D brain CT-Angiography (CTA) in large vessel occlusion (LVO) detection and 3D brain MRI for brain age prediction.

Main Methods:

  • L3P utilizes anisotropic convolutions and unidirectional max pooling to achieve efficient 3D-to-2D information compression.
  • The module was applied to LVO detection (hemisphere classification, presence/absence) and brain age prediction tasks, comparing performance against 2D and fully 3D models.
  • Generalizability was tested using multi-site LVO detection data and a separate T1 MRI dataset for brain age prediction.

Main Results:

  • L3P models achieved performance comparable to stroke-specific 3D models for LVO detection, outperforming standard 2D models while using fewer parameters.
  • The LVO-affected hemisphere detection task yielded an AUC of 0.83 on a large, multi-site test set.
  • On brain age prediction, L3P performed comparably or better than a fully 3D network, demonstrating versatility across tasks and modalities.
  • L3P models also produced more interpretable feature maps.

Conclusions:

  • L3P offers an efficient and versatile approach for analyzing 3D neuroimaging data, enabling high performance in tasks like LVO detection and brain age prediction.
  • The method effectively compresses 3D information into 2D feature maps, reducing computational demands and enhancing model interpretability.
  • L3P shows significant potential for clinical applications, offering a resource-efficient alternative to purely 3D CNNs.