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: Apr 8, 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

3.7K

Dynamic thresholding and robust contrastive techniques for enhanced semi-supervised cardiac segmentation.

Yafei Mi1, Jie Zhang2,3, Hui Jin4

  • 1Department of Cardiology, Taizhou Hospital of Zhejiang Province affiliated to Wenzhou Medical University, Linhai, Zhejiang, China.

Plos One
|April 6, 2026
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

Levosimendan: A New Therapeutical Strategy in Patients with Renal Insufficiency.

Cardiovascular drugs and therapy·2024
Same author

Optimizing storage on fog computing edge servers: A recent algorithm design with minimal interference.

PloS one·2024
Same author

DAFuzz: data-aware fuzzing of in-memory data stores.

PeerJ. Computer science·2023
Same author

Emission reduction effects of vertical environmental regulation: Capacity transfer or energy intensity reduction? Evidence from a quasi-natural experiment in China.

Journal of environmental management·2022
Same author

The Influence Mechanism of Learning Orientation on New Venture Performance: The Chain-Mediating Effect of Absorptive Capacity and Innovation Capacity.

Frontiers in psychology·2022
Same author

Assessing Consumer Preference for Overpackaging Solutions in E-Commerce.

International journal of environmental research and public health·2021
This summary is machine-generated.

This study introduces a new semi-supervised learning method for cardiac segmentation, reducing the need for extensive manual annotations. The approach effectively uses unlabeled data for improved cardiovascular disease diagnosis.

Area of Science:

  • Medical imaging analysis
  • Machine learning for healthcare
  • Cardiovascular disease diagnostics

Background:

  • Manual cardiac segmentation is time-consuming and requires expert annotators.
  • Limited labeled data hinders the training of accurate cardiac segmentation models.
  • Accurate segmentation is vital for diagnosing cardiovascular diseases.

Purpose of the Study:

  • To develop a semi-supervised cardiac segmentation framework utilizing limited labeled data and abundant unlabeled data.
  • To improve the efficiency and accuracy of cardiac structure segmentation.
  • To address the challenge of limited annotated datasets in medical imaging.

Main Methods:

  • Proposed a novel semi-supervised framework for cardiac segmentation.
  • Introduced dynamic pseudo-label threshold maps (pixel-wise, class-wise, adaptive) for robust entropy minimization.

More Related Videos

Automated Joint Space Detection Improves Bone Segmentation Accuracy
06:45

Automated Joint Space Detection Improves Bone Segmentation Accuracy

Published on: November 28, 2025

283

Related Experiment Videos

Last Updated: Apr 8, 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

3.7K
Automated Joint Space Detection Improves Bone Segmentation Accuracy
06:45

Automated Joint Space Detection Improves Bone Segmentation Accuracy

Published on: November 28, 2025

283
  • Incorporated contrastive consistency loss for regularization using unlabeled data.

    • Main Results:

    • Achieved competitive performance against state-of-the-art methods across various labeled data ratios on ACDC and MMWHS datasets.
    • Demonstrated the effectiveness and robustness of individual components through ablation studies.
    • Showcased strong potential for accurate diagnosis with limited annotations.

    Conclusions:

    • The proposed semi-supervised framework significantly enhances cardiac segmentation accuracy with minimal labeled data.
    • The method offers a practical solution for leveraging unlabeled medical data in segmentation tasks.
    • The publicly available code facilitates further research and clinical application.