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: May 20, 2026

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

Domain-agnostic Unsupervised Domain Adaptation Segmentation from 3D Carotid Artery Ultrasound Image.

Zheng Yue, Liping Liu, J David Spence

    IEEE Journal of Biomedical and Health Informatics
    |May 18, 2026
    PubMed
    Summary

    We developed a new method for segmenting 3D carotid artery ultrasound images, improving accuracy even with varied imaging data. This Domain-agnostic Unsupervised Domain Adaptation Segmentation (DaUDASeg) enhances cardiovascular disease risk prediction.

    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

    Deep learning for synovial volume segmentation of the first carpometacarpal joint in osteoarthritis patients.

    Osteoarthritis imaging·2026
    Same author

    Deep Learning-based Segmentation for Assessment of Kidney Tumour Ablation Therapy in CT Images.

    IEEE journal of biomedical and health informatics·2026
    Same author

    A Universal Multimodal Female Pelvic Phantom With Brachytherapy Applications.

    Practical radiation oncology·2026
    Same author

    Attention-driven framework to segment renal ablation zone in posttreatment CT images: a step toward ablation margin evaluation.

    Journal of medical imaging (Bellingham, Wash.)·2026
    Same author

    An Automated Mechatronic System for Registration of Three-Dimensional Ultrasound Images in Cervical Brachytherapy Procedures.

    IEEE transactions on bio-medical engineering·2026
    Same author

    VWV-SSL: Carotid vessel-wall-volume segmentation via sequence structural similarity and augmentation consistency-based self-supervised learning.

    IEEE journal of biomedical and health informatics·2025

    Area of Science:

    • Medical Imaging
    • Artificial Intelligence
    • Cardiovascular Health

    Background:

    • Accurate segmentation of 3D carotid artery (CA) ultrasound (US) images is crucial for cardiovascular disease risk prediction.
    • Unsupervised Domain Adaptation Segmentation (UDASeg) addresses challenges in US image segmentation due to varying imaging parameters.
    • Existing UDASeg methods struggle with domain-agnostic target datasets (DaTD), which lack domain labels and exhibit greater domain shift.

    Purpose of the Study:

    • To propose a novel Domain-agnostic UDASeg (DaUDASeg) method to overcome domain shift in 3D CA US image segmentation from DaTD.
    • To enhance the robustness and accuracy of vessel wall segmentation in diverse 3D CA US datasets.

    Main Methods:

    • Introduced a Domain-agnostic Target Domain Aligning Module (DaTDAM) using a Domain agnostic Style Transfer Model and Structure Anchor Learning to reduce image appearance domain shift.

    More Related Videos

    Manual Segmentation of the Human Choroid Plexus Using Brain MRI
    04:25

    Manual Segmentation of the Human Choroid Plexus Using Brain MRI

    Published on: December 15, 2023

    Related Experiment Videos

    Last Updated: May 20, 2026

    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

    Manual Segmentation of the Human Choroid Plexus Using Brain MRI
    04:25

    Manual Segmentation of the Human Choroid Plexus Using Brain MRI

    Published on: December 15, 2023

  • Implemented Cross Pseudo Labeled Supervision (CPLS) with a similarity-filtered Pseudo Labeled Target Dataset to mitigate source dataset overfitting and provide target data supervision.
  • Trained the DaUDASeg model on a domain-agnostic 3D CA US image dataset.

    • Main Results:

    • The proposed DaUDASeg method significantly outperformed competing methods on a domain-agnostic 3D CA US image dataset.
    • Achieved superior performance with improvements of 3.0% in DSC, 0.09 mm in ASSD, and 0.57 mm in HD95 metrics.
    • Demonstrated enhanced segmentation accuracy and robustness in challenging, domain-agnostic scenarios.

    Conclusions:

    • The developed DaUDASeg method effectively addresses the domain shift problem in 3D CA US image segmentation, particularly for domain-agnostic datasets.
    • The combination of DaTDAM and CPLS offers a robust solution for improving segmentation performance in real-world, varied imaging conditions.
    • This advancement holds significant potential for more accurate cardiovascular disease risk assessment and treatment monitoring.