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 10, 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

3.3K

Cyclic Contrastive Representation Learning for Incomplete Multi-Modal Medical Image Segmentation.

Shihuan He, Zongbao Yang, Jianbo Zhao

    IEEE Journal of Biomedical and Health Informatics
    |November 26, 2025
    PubMed
    Summary

    Cyclic Contrastive Latent Representation Segmentation (CLRS) enhances 3D medical image segmentation accuracy despite missing data. This novel framework effectively models relationships between imaging modalities for robust performance, even with severe data absence.

    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

    Error analysis of Cm measurement under the whole-cell patch-clamp recording.

    Journal of neuroscience methods·2009
    Same author

    Understanding the self-assembly of charged nanoparticles at the water/oil interface.

    Physical chemistry chemical physics : PCCP·2009
    Same author

    [Development of new SSR markers from EST of SSH cDNA libraries on rose fragrance].

    Yi chuan = Hereditas·2009
    Same author

    Crocin and geniposide profiles and radical scavenging activity of gardenia fruits (Gardenia jasminoides Ellis) from different cultivars and at the various stages of maturation.

    Fitoterapia·2009
    Same author

    Small-molecule screening using a human primary cell model of HIV latency identifies compounds that reverse latency without cellular activation.

    The Journal of clinical investigation·2009
    Same author

    Berberine lowers blood glucose in type 2 diabetes mellitus patients through increasing insulin receptor expression.

    Metabolism: clinical and experimental·2009

    Area of Science:

    • Medical Imaging
    • Artificial Intelligence
    • Computer Vision

    Background:

    • Accurate segmentation of multimodal medical images is crucial but challenged by missing data in clinical practice.
    • Incomplete data, particularly missing modalities, significantly degrades segmentation performance.
    • Existing methods struggle with severe modality absence, necessitating robust solutions.

    Purpose of the Study:

    • To develop a joint representation learning framework for robust 3D medical image segmentation under missing modality conditions.
    • To address the performance degradation caused by absent modality-specific information.
    • To improve segmentation accuracy and robustness in scenarios with incomplete multimodal data.

    Main Methods:

    • Proposed Cyclic Contrastive Latent Representation Segmentation (CLRS), a framework incorporating cyclic representation generation and contrastive feature alignment.

    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

    727

    Related Experiment Videos

    Last Updated: Jan 10, 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

    3.3K
    Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
    04:48

    Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

    Published on: July 5, 2024

    727
  • Utilized a unified encoder for feature extraction from available modalities.
  • Employed a synthesis strategy to generate missing latent representations and a channel-wise attention mechanism to enhance modality-specific features.
  • Implemented modality-specific contrastive learning for cross-modal discrimination and disentanglement of shared patterns.

    • Main Results:

    • CLRS demonstrated superior performance on three 3D multimodal datasets, especially in severe missing modality scenarios.
    • Achieved significant improvements over state-of-the-art (SOTA) on the ProstateZS dataset with only a single modality available.
    • Specifically, improvements of over 4.06% for peripheral zone and 2.20% for central gland were observed.

    Conclusions:

    • CLRS offers a robust solution for 3D medical image segmentation with missing modalities.
    • The framework effectively models inter-modality relationships and enhances segmentation robustness.
    • CLRS significantly advances the state-of-the-art in handling incomplete multimodal medical imaging data.