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 Experiment Video

Updated: Jan 11, 2026

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

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

727

Semi-supervised abdominal multi-organ segmentation via dual-task de-biased consistency.

Lina Chen1,2, Xinchi Ye1, Yiqiu Tong3

  • 1School of Computer Science and Technology, Zhejiang Normal University, 688 Yingbin Road, Jinhua, 321004 Zhejiang China.

Health Information Science and Systems
|November 18, 2025
PubMed
Summary

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

Spatial distribution of atmospheric PAHs and their genotoxicity in petrochemical industrialized Lanzhou valley, northwest China.

Environmental science and pollution research international·2017
Same author

Structural formation and charge storage mechanisms for intercalated two-dimensional carbides MXenes.

Physical chemistry chemical physics : PCCP·2017
Same author

Three-dimensional porous ZnCo(2)O(4) sheet array coated with Ni(OH)(2) for high-performance asymmetric supercapacitor.

Journal of colloid and interface science·2017
Same author

Spatiotemporal expression of Wnt3a during striated muscle complex development in rat embryos with ethylenethiourea-induced anorectal malformations.

Molecular medicine reports·2017
Same author

Characterization of Schistosoma japonicum CP1412 protein as a novel member of the ribonuclease T2 molecule family with immune regulatory function.

Parasites & vectors·2017
Same author

Detection of Atmospheric Methyl Mercaptan Using Wavelength Modulation Spectroscopy with Multicomponent Spectral Fitting.

Sensors (Basel, Switzerland)·2017

This study introduces a novel semi-supervised framework for abdominal organ segmentation, improving accuracy for small and difficult classes. The dual-task de-bias approach enhances segmentation performance, particularly in challenging medical imaging scenarios.

Area of Science:

  • Medical Imaging
  • Computer Vision
  • Machine Learning

Background:

  • Abdominal organ segmentation is crucial for medical diagnosis and treatment planning.
  • Existing methods struggle with class imbalance and dynamic organ learning, impacting segmentation accuracy.
  • Semi-supervised learning offers a promising approach to leverage limited labeled data.

Purpose of the Study:

  • To develop a dual-task de-bias consistency semi-supervised framework for abdominal multi-organ segmentation.
  • To address challenges of unbalanced classes and difficult learning of dynamic organs.
  • To improve segmentation accuracy, especially for smaller and less distinct organs.

Main Methods:

  • Proposed a multi-class Hausdorff distance loss for unsupervised learning, enhancing sensitivity to shapes and boundaries.
Keywords:
Abdominal organ segmentationClass imbalanceDouble debiasingThe Hausdorff distance loss function

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

3.3K

Related Experiment Videos

Last Updated: Jan 11, 2026

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

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

727
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

3.3K
  • Implemented a dual debiasing strategy for pixel and contour prediction tasks to dynamically adjust for data and learning biases.
  • Employed a semi-supervised learning framework to utilize limited labeled data effectively.

    • Main Results:

    • The proposed framework achieved optimal performance on two distinct datasets.
    • Demonstrated significant improvement in the accuracy of segmenting small and difficult abdominal classes.
    • Achieved a 6.9% improvement in average Dice Similarity Coefficient (DSC) on the Synapse dataset with only 10% labels.

    Conclusions:

    • The dual-task de-bias consistency semi-supervised framework effectively enhances abdominal multi-organ segmentation.
    • The novel loss function and debiasing strategy are key to improving performance on challenging cases.
    • This approach shows great potential for improving automated medical image analysis in clinical settings.