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 28, 2026

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

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

IPSM-UNet: An Inverted Pyramid-Shaped U-Net++ Architecture with Multi-Resolution Information Interaction for Coronary

Yinong Liao1,2, Wei Li3,4, Guopeng Liu4

  • 1State Key Laboratory of Multimodal Artificial Intelligence, Institute of Automation, Chinese Academy of Sciences, Beijing 100190, China.

Journal of Imaging
|May 26, 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

Structural Evolution and Coordination Behavior of Ta<sup>5+</sup> in Binary Li<sub>2</sub>O-Ta<sub>2</sub>O<sub>5</sub> Melts: <i>In Situ</i> High-Temperature Raman and High-Energy X-ray Diffraction Studies.

Inorganic chemistry·2026
Same author

In Situ High-Temperature Raman Spectroscopy and Simulations for Correlating Structure Evolution with Viscosity in the Binary Potassium Titanate System.

Analytical chemistry·2025
Same author

Engineered mRNAs With Stable Structures Minimize Double-stranded RNA Formation and Increase Protein Expression.

Journal of molecular biology·2024
Same author

Efficient circularization of protein-encoding RNAs via a novel cis-splicing system.

Nucleic acids research·2024
Same author

NanoMUD: Profiling of pseudouridine and N1-methylpseudouridine using Oxford Nanopore direct RNA sequencing.

International journal of biological macromolecules·2024
Same author

Cyclo[2]pyridine[8]pyrrole: An Expanded Porphyrinoid with Three Closed-Shell Oxidation States.

Journal of the American Chemical Society·2024

A novel Inverted Pyramid-Shaped Multi-resolution U-Net (IPSM-UNet) improves coronary artery segmentation accuracy. This deep learning approach enhances vessel continuity and quality, aiding diagnosis and intervention planning.

Area of Science:

  • Medical Imaging
  • Deep Learning
  • Cardiovascular Imaging

Background:

  • Accurate coronary artery segmentation is critical for diagnosing cardiovascular diseases and planning interventions.
  • Conventional U-Net architectures struggle with thin, low-contrast vessels and maintaining segmentation continuity.

Purpose of the Study:

  • To introduce a novel deep learning model, the Inverted Pyramid-Shaped Multi-resolution U-Net (IPSM-UNet), for enhanced coronary artery segmentation.
  • To address limitations of existing methods in segmenting thin, low-contrast vessels and preserving vessel continuity.

Main Methods:

  • Developed IPSM-UNet, a dual U-Net++ architecture incorporating multi-resolution feature interaction and aggregation.
  • Implemented layer-wise deep supervision to refine segmentation accuracy.
Keywords:
IPSM-UNetU-Net++coronary artery segmentationdeep supervisionfeature aggregation

Related Experiment Videos

Last Updated: May 28, 2026

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

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

  • Evaluated the model on multiple public datasets (DRIVE, CHASE_DB1, DCA1) and an internal coronary angiography dataset.

    • Main Results:

    • IPSM-UNet demonstrated competitive or superior performance across all evaluated datasets.
    • Achieved high segmentation metrics, including F1 scores up to 0.8590 and accuracy up to 0.9879.
    • Significantly improved segmentation quality, particularly for small-caliber vessels, and enhanced vessel continuity.

    Conclusions:

    • IPSM-UNet offers a robust solution for accurate coronary artery segmentation.
    • The model's improvements in continuity and detail support advanced coronary analysis and clinical applications.
    • This architecture advances deep learning applications in cardiovascular imaging.