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

Generative restoration for cardiac PET attenuation correction: a two-stage 3D DDIM framework optimizing fidelity and clinical controllability.

EJNMMI physics·2026
Same author

Model-unrolled fast MRI with weakly supervised lesion enhancement.

Medical image analysis·2025
Same author

Collaborative and privacy-preserving cross-vendor united diagnostic imaging via server-rotating federated machine learning.

Communications engineering·2025
Same author

Large-vocabulary forensic pathological analyses via prototypical cross-modal contrastive learning.

Nature communications·2025
Same author

Anatomy-Aware Deep Unrolling for Task-Oriented Acceleration of Multi-Contrast MRI.

IEEE transactions on medical imaging·2025
Same author

Explainable AI for sharp injury identification using transfer learning with pre-trained deep neural networks.

Forensic science international·2025
Same journal

ContiMorph: An unsupervised learning framework for cardiac motion tracking with time-continuous diffeomorphism.

Medical image analysis·2026
Same journal

MedP-CLIP: Medical CLIP with region-aware prompt integration.

Medical image analysis·2026
Same journal

Multi-organ guided diagnosis of mild cognitive impairment via hierarchical alignment and knowledge distillation.

Medical image analysis·2026
Same journal

SUDA: Simultaneous unsupervised knowledge distillation and adaptation of foundation models for efficient pathological image analysis.

Medical image analysis·2026
Same journal

Beyond the LUMIR challenge: The pathway to foundational registration models.

Medical image analysis·2026
Same journal

Annotation-efficient medical image segmentation via cross-latent graphs and vector-quantized memory.

Medical image analysis·2026
See all related articles

Related Experiment Video

Updated: Nov 6, 2025

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

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

607

Diverse data augmentation for learning image segmentation with cross-modality annotations.

Xu Chen1, Chunfeng Lian1, Li Wang1

  • 1Department of Radiology and Biomedical Research Imaging Center (BRIC), University of North Carolina, Chapel Hill, NC, USA.

Medical Image Analysis
|May 6, 2021
PubMed
Summary
This summary is machine-generated.

Annotated medical image data is scarce. A new Diverse Data Augmentation Generative Adversarial Network (DDA-GAN) method uses unpaired images to train segmentation models, reducing annotation needs for medical imaging.

Keywords:
Data augmentationDisentangled representation learningGenerative adversarial learningMedical image 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

3.1K
From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data
12:08

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data

Published on: August 13, 2014

24.8K

Related Experiment Videos

Last Updated: Nov 6, 2025

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

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

607
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

3.1K
From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data
12:08

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data

Published on: August 13, 2014

24.8K

Area of Science:

  • Medical image analysis
  • Computer vision
  • Machine learning

Background:

  • Reliable medical image segmentation models require extensive annotated data, which is difficult and time-consuming to obtain.
  • Variability in manual annotations across different imaging modalities poses a significant challenge.
  • Leveraging annotation-rich source modalities can help train models for annotation-poor target modalities.

Purpose of the Study:

  • To introduce a novel Diverse Data Augmentation Generative Adversarial Network (DDA-GAN) for training image segmentation models.
  • To address the challenge of limited annotated data in target domains by utilizing information from annotated source domains.
  • To develop a method that reduces the need for manual annotation in medical image segmentation.

Main Methods:

  • The DDA-GAN employs one-to-many source-to-target translation to generate diverse augmented data for the target domain.
  • It uses unpaired images from source and target domains in an end-to-end convolutional neural network architecture.
  • The network disentangles domain-invariant structural features from domain-specific appearance features, combining them for augmentation.

Main Results:

  • The DDA-GAN successfully trains segmentation models for unannotated target domains by leveraging annotated source domains.
  • Qualitative and quantitative evaluations demonstrate the method's effectiveness compared to state-of-the-art approaches.
  • The approach is validated for segmenting craniomaxillofacial bony structures in MRI and cardiac substructures in CT.

Conclusions:

  • The DDA-GAN effectively mitigates the need for extensive manual annotations in medical image segmentation.
  • This method enables the development of robust segmentation models even with limited or no annotated data in the target domain.
  • The DDA-GAN offers a promising solution for improving medical image analysis workflows across various modalities.