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: Jul 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

LSR-Diff: A Diffusion Model Synthesizing Level Set Representations for Reliable Segmentation of Medical Images with

Wenbo Gao, Haoyu Cao, James Chung-Wai Cheung

    IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society
    |July 8, 2026
    PubMed
    Summary
    This summary is machine-generated.

    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

    Integrating clinical, proteomics, and polygenic scores to improve cardiovascular risk prediction: a prospective cohort study.

    Journal of advanced research·2026
    Same author

    Efficacy and Safety of Ciprofol Versus Propofol for Anesthesia in Older Patients: A Systematic Review and Meta-Analysis of Randomized Controlled Trials.

    Drugs & aging·2026
    Same author

    Gut Microbiota and Metabolite Remodeling Underlies the Anxiolytic Effect of <i>Anshen Bunao</i> Oral Liquid.

    Pharmaceuticals (Basel, Switzerland)·2026
    Same author

    Extraction, purification, bioactivities, and applications of active polysaccharides (AGP) from Ampelopsis grossedentata: A review.

    Carbohydrate research·2026
    Same author

    Mechanisms and biomarkers of immune checkpoint inhibitor-associated myocarditis: from T cell imbalance to multicellular crosstalk.

    Frontiers in immunology·2026
    Same author

    Dietary patterns and knowledge-attitude-practice factors are associated with late recurrence of non-muscle-invasive bladder cancer: a case-control study.

    Frontiers in nutrition·2026

    This study introduces a novel diffusion model for medical image segmentation, improving boundary accuracy and topological consistency. The new method addresses limitations of existing models, leading to more precise disease diagnosis and analysis.

    Area of Science:

    • Medical Imaging
    • Artificial Intelligence
    • Computer Vision

    Background:

    • Accurate boundary segmentation is crucial for medical diagnosis but challenging due to complex image data.
    • Diffusion Probabilistic Models (DPMs) offer improved boundary delineation but struggle with discrete mask representations.
    • Existing aggregation methods for DPMs often disregard structural consistency, leading to artifacts.

    Purpose of the Study:

    • To develop a novel diffusion model for enhanced medical image segmentation.
    • To improve boundary delineation and topological accuracy in segmentation tasks.
    • To address the limitations of discrete mask learning and arbitrary aggregation in DPMs.

    Main Methods:

    • Proposed the Level Set Representation Diffusion model (LSR-Diff) with a hybrid mask representation.

    Related Experiment Videos

    Last Updated: Jul 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

  • Introduced Ensemble Aggregation via Level Set Evolution (EASE) for robust prediction merging.
  • Utilized ambiguity estimation and anatomical structure guidance within the EASE module.

    • Main Results:

    • LSR-Diff demonstrated competitive overall performance across diverse clinical applications.
    • The proposed method significantly improved edge quality and topological accuracy.
    • EASE module effectively refined boundary topology, preventing mask assembly issues.

    Conclusions:

    • LSR-Diff offers a robust approach to medical image segmentation, enhancing boundary precision.
    • The hybrid mask representation and EASE strategy effectively address DPM limitations.
    • This work advances segmentation accuracy for critical clinical applications.