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

Force Classification01:22

Force Classification

2.8K
Forces play a crucial role in the study of physics and engineering. They are essential in describing the motion, behavior, and equilibrium of objects in the physical world. Forces can be classified based on their origin, type, and direction of action.
Contact and non-contact forces are two of the most widely used categories of forces. As the name suggests, contact forces require physical contact between two objects to act upon each other. Examples of contact forces include frictional,...
2.8K

You might also read

Related Articles

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

Sort by
Same author

Fluid Volumes Longitudinal Modeling to Predict Atrophy and Fibrosis in Neovascular Age-Related Macular Degeneration.

Ophthalmology science·2026
Same author

STAGE challenge: Structural-Functional Transition in Glaucoma Assessment.

Medical image analysis·2026
Same author

Choriocapillaris Flow-Enriched Prediction of Retinal Sensitivity Using OCT-Derived Biomarkers in Intermediate Age-Related Macular Degeneration.

Journal of clinical medicine·2026
Same author

I-SCREEN: Development of an AI-based infrastructure for community-wide screening and prediction of progression in age-related macular degeneration providing accessible shared care.

Eye (London, England)·2026
Same author

OCT-based AI-assisted phenotyping of intermediate AMD in the prospective PINNACLE trial: PINNACLE Study Report 9.

The British journal of ophthalmology·2026
Same author

Choriocapillaris Flow Deficits Are Spatially Linked to Retinal Pigment Epithelium and Photoreceptor Integrity in Geographic Atrophy.

Investigative ophthalmology & visual science·2026

Related Experiment Video

Updated: May 5, 2026

Doppler Optical Coherence Tomography of Retinal Circulation
10:46

Doppler Optical Coherence Tomography of Retinal Circulation

Published on: September 18, 2012

18.7K

VLFATRollout: Fully transformer-based classifier for retinal OCT volumes.

Marzieh Oghbaie1, Teresa Araújo1, Ursula Schmidt-Erfurth2

  • 1Christian Doppler Laboratory for Artificial Intelligence in Retina, Department of Ophthalmology and Optometry, Medical University of Vienna, Austria; Institute of Artificial Intelligence, Center for Medical Data Science, Medical University of Vienna, Austria.

Computerized Medical Imaging and Graphics : the Official Journal of the Computerized Medical Imaging Society
|November 3, 2024
PubMed
Summary

This study introduces VLFATRollout, a transformer-based framework for 3D medical image analysis. It improves classification accuracy on retinal OCT scans by efficiently processing variable-resolution volumes and focusing on relevant features.

Keywords:
3D volume classificationExplainabilityOptical coherence tomographyTransformers

More Related Videos

Retinal Vascular Reactivity as Assessed by Optical Coherence Tomography Angiography
07:23

Retinal Vascular Reactivity as Assessed by Optical Coherence Tomography Angiography

Published on: March 26, 2020

7.1K
A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
04:23

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

1.7K

Related Experiment Videos

Last Updated: May 5, 2026

Doppler Optical Coherence Tomography of Retinal Circulation
10:46

Doppler Optical Coherence Tomography of Retinal Circulation

Published on: September 18, 2012

18.7K
Retinal Vascular Reactivity as Assessed by Optical Coherence Tomography Angiography
07:23

Retinal Vascular Reactivity as Assessed by Optical Coherence Tomography Angiography

Published on: March 26, 2020

7.1K
A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
04:23

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

1.7K

Area of Science:

  • Medical Image Analysis
  • Artificial Intelligence
  • Computer Vision

Background:

  • 3D transformer architectures show promise in video analysis but face challenges with high-resolution 3D medical volumes.
  • Limitations include reduced efficiency due to numerous 3D patches and background noise distracting attention mechanisms.
  • Variability in slice count per volume complicates processing of arbitrary resolutions, risking loss of diagnostic detail with subsampling.

Purpose of the Study:

  • To introduce an end-to-end transformer-based framework, VLFATRollout, for efficient and accurate classification of volumetric medical data.
  • To address the challenges of high-resolution 3D medical volumes, including patch count, background noise, and variable slice numbers.
  • To enhance the learning capacity and generalization of models for processing medical volumes of any resolution.

Main Methods:

  • Developed VLFATRollout, an end-to-end transformer framework for volumetric data classification.
  • Utilized transformer attention matrices to mine slice-level foreground-background information.
  • Employed randomization of volume-wise resolution during training to improve the generalization of learnable positional embeddings.

Main Results:

  • VLFATRollout achieved an average improvement of 5.47% in balanced accuracy on retinal optical coherence tomography (OCT) volume classification.
  • Outperformed leading convolutional models in a 5-class diagnostic task.
  • Demonstrated effectiveness in enhancing slice-level representation and adaptability to different volume resolutions.

Conclusions:

  • VLFATRollout offers an effective solution for transformer-based medical image analysis, particularly for high-resolution 3D volumes.
  • The framework enhances diagnostic accuracy and handles variable volume resolutions, addressing key limitations of current transformer applications.
  • The study paves the way for advanced transformer applications in medical imaging, with code available for reproducibility.