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: Nov 24, 2025

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

Annotation-Efficient Learning for Medical Image Segmentation Based on Noisy Pseudo Labels and Adversarial Learning.

Lu Wang, Dong Guo, Guotai Wang

    IEEE Transactions on Medical Imaging
    |December 28, 2020
    PubMed
    Summary

    This study introduces an annotation-efficient framework for medical image segmentation using generative adversarial networks (GANs). The method achieves segmentation performance comparable to manual annotation without requiring training image labels.

    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 effects and mechanisms of modified Xiaoyaosan on chronic unpredictable mild stress (CUMS)-induced depressive mice based on network pharmacology.

    Journal of ethnopharmacology·2024
    Same author

    The relationship between Listeria infections and host immune responses: Listeriolysin O as a potential target.

    Biomedicine & pharmacotherapy = Biomedecine & pharmacotherapie·2024
    Same author

    Dynamic coordination engineering of 2D PhenPtCl<sub>2</sub> nanosheets for superior hydrogen evolution.

    Nature communications·2024
    Same author

    Mesoporous Polyimide Thin Films as Dendrite-Suppressing Separators for Lithium-Metal Batteries.

    ACS nano·2023
    Same author

    Acupuncture for vitiligo: An overview of systematic reviews and meta-analysis.

    Journal of cosmetic dermatology·2023
    Same author

    AMPK-HIF-1α signaling enhances glucose-derived de novo serine biosynthesis to promote glioblastoma growth.

    Journal of experimental & clinical cancer research : CR·2023

    Area of Science:

    • Medical Imaging
    • Artificial Intelligence
    • Computer Vision

    Background:

    • Deep learning excels in medical image segmentation but requires extensive manual annotations, which are costly and time-consuming.
    • Annotation-efficient methods are crucial for advancing AI in healthcare by reducing data acquisition burdens.

    Purpose of the Study:

    • To develop an annotation-efficient learning framework for medical image segmentation.
    • To leverage unpaired medical images and auxiliary masks for training segmentation models.
    • To achieve segmentation performance comparable to supervised methods without manual annotations.

    Main Methods:

    • Utilized an improved Cycle-Consistent Generative Adversarial Network (GAN) for learning from unpaired medical images.
    • Employed a Variational Auto-encoder (VAE)-based discriminator with auxiliary masks to generate pseudo labels under shape constraints.

    More Related Videos

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

    Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

    Published on: July 5, 2024

    655

    Related Experiment Videos

    Last Updated: Nov 24, 2025

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

    Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

    Published on: July 5, 2024

    655
  • Introduced a Discriminator-guided Generator Channel Calibration (DGCC) module for generator refinement.
  • Implemented a noise-robust iterative learning method with noise-weighted Dice loss to handle pseudo-label inaccuracies.

    • Main Results:

    • The VAE-based discriminator and DGCC module effectively produced high-quality pseudo labels.
    • The noise-robust learning strategy successfully mitigated the impact of noisy pseudo labels.
    • Segmentation performance using the proposed framework closely matched or surpassed results from methods relying on human annotations.

    Conclusions:

    • The developed framework offers a viable solution for annotation-efficient medical image segmentation.
    • The approach demonstrates the potential to significantly reduce the need for manual annotations in medical AI.
    • This method holds promise for broader applications in medical image analysis where labeled data is scarce.