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

Phosphorylation of WDR48 by phototropins drives starch degradation to promote stomatal opening.

Nature communications·2026
Same author

High-Throughput Optimization of a High-Pressure Catalytic Reaction.

Journal of chemical information and modeling·2025
Same author

Delocalized, asynchronous, closed-loop discovery of organic laser emitters.

Science (New York, N.Y.)·2024
Same author

Automatic enhancement preprocessing for segmentation of low quality cell images.

Scientific reports·2024
Same author

Creation of synthetic contrast-enhanced computed tomography images using deep neural networks to screen for renal cell carcinoma.

Nagoya journal of medical science·2023
Same author

Adaptive t-vMF dice loss: An effective expansion of dice loss for medical image segmentation.

Computers in biology and medicine·2023

Related Experiment Video

Updated: Jul 8, 2025

Analysis of Tubular Membrane Networks in Cardiac Myocytes from Atria and Ventricles
10:30

Analysis of Tubular Membrane Networks in Cardiac Myocytes from Atria and Ventricles

Published on: October 15, 2014

20.6K

Expanded tube attention for tubular structure segmentation.

Sota Kato1, Kazuhiro Hotta2

  • 1Department of Electrical, Information, Materials and Materials Engineering, Meijo University, Tempaku-ku, Nagoya, Aichi, 468-8502, Japan. 150442030@ccalumni.meijo-u.ac.jp.

International Journal of Computer Assisted Radiology and Surgery
|December 19, 2023
PubMed
Summary

This study introduces a new method for segmenting thin tubular structures in images. The expanded tube attention (ETA) module improves accuracy by first learning from thickened pseudo-labels and then refining to original structures.

Keywords:
Attention mechanismMorphological transformationPseudo-labelTubular structure segmentation

More Related Videos

Trophoblast Cell Recovery from Angiogenesis-Tube Formation Assay for Differentiation Marker Expression Analysis
05:30

Trophoblast Cell Recovery from Angiogenesis-Tube Formation Assay for Differentiation Marker Expression Analysis

Published on: November 8, 2024

481
Detection and Quantification of Tunneling Nanotubes Using 3D Volume View Images
12:45

Detection and Quantification of Tunneling Nanotubes Using 3D Volume View Images

Published on: August 31, 2022

2.9K

Related Experiment Videos

Last Updated: Jul 8, 2025

Analysis of Tubular Membrane Networks in Cardiac Myocytes from Atria and Ventricles
10:30

Analysis of Tubular Membrane Networks in Cardiac Myocytes from Atria and Ventricles

Published on: October 15, 2014

20.6K
Trophoblast Cell Recovery from Angiogenesis-Tube Formation Assay for Differentiation Marker Expression Analysis
05:30

Trophoblast Cell Recovery from Angiogenesis-Tube Formation Assay for Differentiation Marker Expression Analysis

Published on: November 8, 2024

481
Detection and Quantification of Tunneling Nanotubes Using 3D Volume View Images
12:45

Detection and Quantification of Tunneling Nanotubes Using 3D Volume View Images

Published on: August 31, 2022

2.9K

Area of Science:

  • Medical Image Analysis
  • Computer Vision
  • Biomedical Engineering

Background:

  • Semantic segmentation of thin tubular structures like blood vessels is challenging due to extreme pixel imbalance.
  • Existing methods often fail, breaking predicted regions because ground truth is very thin.

Purpose of the Study:

  • To develop a novel training method for improving semantic segmentation of tubular structures.
  • To introduce an attention module that addresses the challenges of thin and unbalanced data.

Main Methods:

  • A new training approach using pseudo-labels generated via morphological transformation.
  • An attention module, the expanded tube attention (ETA) module, utilizing thickened pseudo-labels.
  • A progressive learning strategy where the network first learns from thickened regions and then refines to original thin regions.

Main Results:

  • Experiments on retina vessel image datasets demonstrated improved clDice metric accuracy.
  • The proposed method using ETA modules outperformed conventional segmentation techniques.
  • Validation confirmed the effectiveness of the ETA module in enhancing segmentation performance.

Conclusions:

  • The novel expanded tube attention (ETA) module effectively addresses the challenges of segmenting thin tubular structures.
  • The proposed method facilitates an easy-to-hard learning process, enhancing segmentation accuracy.
  • This approach offers a promising solution for accurate semantic segmentation in biomedical imaging.