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: Sep 11, 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

2.9K

Semi-supervised dual generative adversarial network for low-dose CT artifact suppression.

Shangguan Hong, Huiying Ren, Xiong Zhang

    Optics Express
    |August 13, 2025
    PubMed
    Summary

    This study introduces a semi-supervised dual generative adversarial network for low-dose computed tomography (LDCT) denoising. The method effectively reduces noise and artifacts in LDCT images using both paired and unpaired data.

    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

    Molecular epidemiology and humoral immunity of BK polyomavirus in the general population of Southern China.

    Virology·2026
    Same author

    PIEZO1 mediates orthodontic tension force-induced alveolar bone remodeling via the Ca²⁺/YAP-TAZ pathway in periodontal ligament stem cells.

    Journal of translational medicine·2026
    Same author

    OX40<sup>+</sup> regulatory T cells suppress neoadjuvant chemotherapy response in NSCLC and reversed by combination of PD-1 blockade therapy.

    Lung cancer (Amsterdam, Netherlands)·2026
    Same author

    Construction of Hydrogel Electrolytes With Wide Potential Windows Based on Dimethyl Sulfone-Mediated Interface Regulation Mechanisms.

    Small (Weinheim an der Bergstrasse, Germany)·2026
    Same author

    The digital dividend of the views on aging: how digital engagement influences subjective aging in Chinese older adults.

    Frontiers in public health·2026
    Same author

    Population pharmacokinetics of levofloxacin: a systematic review.

    Drug metabolism reviews·2026

    Area of Science:

    • Medical Imaging
    • Artificial Intelligence
    • Computer Vision

    Background:

    • Deep learning, particularly supervised methods, has advanced low-dose computed tomography (LDCT) denoising.
    • Acquiring sufficient paired training data for LDCT denoising is challenging.
    • Supervised learning is difficult to apply directly to unpaired CT image data.

    Purpose of the Study:

    • To develop a novel semi-supervised deep learning approach for LDCT denoising.
    • To effectively utilize both paired and unpaired CT data for improved network performance.
    • To enhance artifact and noise suppression in LDCT images.

    Main Methods:

    • Designed a semi-supervised dual generative adversarial network (GAN).
    • Employed alternating iterations of supervised and unsupervised training using paired and unpaired data.

    More Related Videos

    Management of Respiratory Motion Artefacts in 18F-fluorodeoxyglucose Positron Emission Tomography using an Amplitude-Based Optimal Respiratory Gating Algorithm
    06:53

    Management of Respiratory Motion Artefacts in 18F-fluorodeoxyglucose Positron Emission Tomography using an Amplitude-Based Optimal Respiratory Gating Algorithm

    Published on: July 23, 2020

    5.7K

    Related Experiment Videos

    Last Updated: Sep 11, 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

    2.9K
    Management of Respiratory Motion Artefacts in 18F-fluorodeoxyglucose Positron Emission Tomography using an Amplitude-Based Optimal Respiratory Gating Algorithm
    06:53

    Management of Respiratory Motion Artefacts in 18F-fluorodeoxyglucose Positron Emission Tomography using an Amplitude-Based Optimal Respiratory Gating Algorithm

    Published on: July 23, 2020

    5.7K
  • Utilized interactive adversarial training between two GANs for image generation and noise reduction.
  • Incorporated a multi-description loss function to improve training stability and performance.

    • Main Results:

    • Achieved effective suppression of artifacts and noise in LDCT images.
    • Demonstrated superior noise reduction compared to existing supervised and unsupervised methods.
    • Preserved the original structural information of CT images effectively.

    Conclusions:

    • The proposed semi-supervised dual GAN effectively denoises LDCT images.
    • This approach overcomes limitations of data acquisition and utilization of unpaired data.
    • It offers a promising method for improving LDCT image quality while maintaining anatomical integrity.