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

The prognostic implications of autoimmune hemolytic anemia in chronic lymphocytic leukemia across treatment eras.

Haematologica·2026
Same author

Extracellular Vesicles for Therapeutic Applications: A Translational Framework Integrating Sources, Administration Routes, Indications, Quality Control, and Regulatory Systems.

International journal of nanomedicine·2026
Same author

Disease characteristics, treatment, and outcomes in Chinese chronic lymphocytic leukemia patients following BTK inhibitor discontinuation: a multicenter real-world study.

Frontiers in medicine·2026
Same author

Single-Cell Profiling Identifies SLC2A5-Mediated Fructose Metabolism as a Vulnerability in Primary CNS Lymphoma.

Advanced science (Weinheim, Baden-Wurttemberg, Germany)·2026
Same author

MYD88 variants in chronic lymphocytic leukemia exhibit distinct biological behaviors and prognostic implications.

Leukemia·2026
Same author

Talquetamab-Daratumumab in Relapsed or Refractory Myeloma.

The New England journal of medicine·2026

Related Experiment Video

Updated: Aug 9, 2025

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

472

MG-Net: Multi-level global-aware network for thymoma segmentation.

Jingyuan Li1, Wenfang Sun2, Karen M von Deneen1

  • 1Center for Brain Imaging, School of Life Science and Technology, Xidian University & Engineering Research Center of Molecular and Neuro Imaging, Ministry of Education, Xi'an, Shaanxi, 710126, China; International Joint Research Center for Advanced Medical Imaging and Intelligent Diagnosis and Treatment & Xi'an Key Laboratory of Intelligent Sensing and Regulation of Trans-Scale Life Information, School of Life Science and Technology, Xidian University, Xi'an, Shaanxi, 710126, China.

Computers in Biology and Medicine
|February 15, 2023
PubMed
Summary

This study introduces a novel deep learning network, MG-Net, for improved thymoma segmentation in CT scans. MG-Net enhances global awareness to accurately identify thymomas of varying characteristics.

Keywords:
Attention mechanismConvolution neural networkMedical imageSelf-attentionSemantic segmentation

More Related Videos

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.9K
Deep Learning-Based Segmentation of Cryo-Electron Tomograms
10:25

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

9.1K

Related Experiment Videos

Last Updated: Aug 9, 2025

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

472
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.9K
Deep Learning-Based Segmentation of Cryo-Electron Tomograms
10:25

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

9.1K

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Computer Vision

Background:

  • Automatic thymoma segmentation in contrast-enhanced computed tomography (CECT) is crucial for diagnosis.
  • Convolutional neural networks (CNNs) struggle with thymoma segmentation due to shape, scale, and texture variations, stemming from their inherent locality.
  • A deep learning approach with enhanced global awareness is needed to address these limitations.

Purpose of the Study:

  • To develop a deep learning network that improves global awareness for accurate thymoma segmentation.
  • To overcome the limitations of traditional CNNs in handling diverse thymoma characteristics.

Main Methods:

  • Proposed a multi-level global-aware network (MG-Net) incorporating multi-level feature interaction and integration.
  • Designed a cross-attention block (CAB) for pixel-wise feature interaction, creating a Global Enhanced Convolution Block.
  • Developed a Global Spatial Attention Module and an Adaptive Attention Fusion Module to enhance semantic consistency and detail preservation.

Main Results:

  • MG-Net demonstrated superior segmentation performance and generalization ability compared to state-of-the-art models.
  • Evaluated on CECT and NIH Pancreas-CT datasets, confirming the effectiveness of all designed components.
  • Both qualitative and quantitative results validated MG-Net's accuracy and generalization.

Conclusions:

  • MG-Net achieves accurate thymoma segmentation through its global-aware capabilities.
  • The network exhibits strong generalization ability across different tasks.
  • The developed model offers a significant advancement in automated medical image segmentation for thymoma diagnosis.