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: Jun 9, 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

SBEM-UNet: A Semantic Boundary and Contour-Enhanced Framework for Semisupervised Medical Image Segmentation.

Hongwei Zhang1, Kaijun Yang2, Meifeng Shi3

  • 1College of Computer Science and Engineering, Chongqing University of Technology, Chongqing, China. zhanghongwei@cqut.edu.cn.

Journal of Imaging Informatics in Medicine
|June 8, 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

Transformer-driven automated analysis of social media narrative structure: An exploration based on sentiment framing and thematic agenda.

PloS one·2026
Same author

Efficient analysis of n-alkanes by high-pressure photoionization-induced NO<sub>2</sub><sup>+</sup> chemical ionization time-of-flight mass spectrometry.

Talanta·2026
Same author

Gender inequality in academic promotion trajectories in Japan.

Scientific reports·2026
Same author

Tucatinib alleviates postmenopausal osteoporosis by suppressing osteoclast differentiation via regulating the DRP1/NFATc1/CTSK signaling pathway.

Biochemical pharmacology·2026
Same author

Study on the Fabrication of Coating-Free Superhydrophobic Aluminum Alloy Surfaces by Femtosecond Laser and Its Wettability Control Mechanism.

Nanomaterials (Basel, Switzerland)·2026
Same author

Kaempferol alleviates T-cell immunosenescence and inflammaging in aged mice via the SIRT3-LKB1-AMPK-mitophagy pathway.

Immunity & ageing : I & A·2026

SBEM-UNet enhances medical image segmentation by improving boundary delineation in low-annotation scenarios. This novel framework effectively addresses blurred edges and tissue overlap for more accurate anatomical segmentation.

Area of Science:

  • Medical Image Analysis
  • Deep Learning
  • Computer Vision

Background:

  • Medical image segmentation faces challenges due to ambiguous boundaries, tissue overlap, and weak gradients.
  • Deep learning models struggle with modeling ambiguous boundaries, especially with limited annotations.

Purpose of the Study:

  • To propose SBEM-UNet, a novel semisupervised learning framework to improve medical image segmentation accuracy.
  • To explicitly address boundary blurring and discontinuity in anatomical delineation.

Main Methods:

  • SBEM-UNet incorporates a semantic boundary enhancement module (SBEM) and a contour enhancement decoder (CED).
  • SBEM uses multiscale semantic aggregation and attention for enhanced structural discriminability.
  • CED performs fine-grained contour modeling using dynamic boundary extraction and adaptive feature modulation.
Keywords:
Boundary ambiguityMedical image segmentationSemantic boundary enhancement moduleSemisupervised learning

Related Experiment Videos

Last Updated: Jun 9, 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

  • The framework utilizes pseudolabel consistency regularization for robustness under limited annotations.

    • Main Results:

    • SBEM-UNet demonstrated superior performance compared to existing semisupervised methods on public benchmarks.
    • The model achieved improved region-based accuracy and boundary delineation quality.
    • Effectiveness was highlighted in low-annotation clinical scenarios.

    Conclusions:

    • SBEM-UNet offers a robust solution for accurate medical image segmentation, particularly when annotations are scarce.
    • The framework effectively enhances the modeling of ambiguous and discontinuous boundaries.
    • It holds significant practical value for clinical applications requiring precise anatomical delineation.