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 Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Soil fungal diversity and plant resilience under the dual stress of microplastics and heavy metals: A comparative study of lettuce and green onion.

Ecotoxicology (London, England)·2026
Same author

Unlocking poly(lactic acid) degradation through biochar-induced environmental shifts and functional consortia enrichment in composting systems.

International journal of biological macromolecules·2026
Same author

Bioaugmentation influences PBAT biodegradation patterns during composting through associated shifts in plastisphere communities and surface properties.

Bioresource technology·2025
Same author

Non-Contact Heart Rate Variability Monitoring with FMCW Radar via a Novel Signal Processing Algorithm.

Sensors (Basel, Switzerland)·2025
Same author

Placido Sub-Pixel Edge Detection Algorithm Based on Enhanced Mexican Hat Wavelet Transform and Improved Zernike Moments.

Journal of imaging·2025
Same author

Microbial mechanisms driving enhanced short-chain fatty acid and alcohol production during food waste fermentation via biochar addition in the pilot-scale reactor.

Environmental research·2025

Related Experiment Video

Updated: May 27, 2025

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

2.6K

Spine X-ray image segmentation based on deep learning and marker controlled watershed.

Yating Xiao1,2, Yan Chen1,2, Yong Zhang3

  • 1School of Information and Communication Engineering, North University of China, Taiyuan, Shanxi Province, China.

Journal of X-Ray Science and Technology
|February 20, 2025
PubMed
Summary

This study introduces a novel deep learning and marker-controlled watershed method for accurate vertebral segmentation in spine X-rays. The approach effectively addresses adjacent vertebral adhesion, improving diagnostic analysis for spinal diseases.

Keywords:
deep learningmarker controlled watershedspine X-ray imagevertebral localizationvertebral segmentation

More Related Videos

Deep Learning-Based Segmentation of Cryo-Electron Tomograms
10:25

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

8.5K
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

349

Related Experiment Videos

Last Updated: May 27, 2025

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

2.6K
Deep Learning-Based Segmentation of Cryo-Electron Tomograms
10:25

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

8.5K
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

349

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Spine Diagnostics

Background:

  • Automatic vertebral segmentation is crucial for objective spine image analysis and diagnosing spinal diseases.
  • Vertebrae present challenges due to inter-class similarity, intra-class variability, and adhesion between adjacent segments.
  • Accurate demarcation of vertebral boundaries is essential for reliable diagnosis.

Purpose of the Study:

  • To develop an image segmentation method that overcomes the challenge of adjacent vertebral adhesion.
  • To ensure precise boundary delineation between adjacent vertebrae in spinal images.
  • To enhance the accuracy of automatic vertebral segmentation for clinical applications.

Main Methods:

  • A dual-path deep learning model combining a localization path (HRNet with bone direction loss) and a segmentation path (VU-Net with position information perception module).
  • Utilized HRNet for vertebral center localization and VU-Net for preliminary segmentation, with HRNet guiding VU-Net.
  • Integrated deep learning outputs to initialize a marker-controlled watershed algorithm for fine segmentation and adhesion suppression.

Main Results:

  • The method was validated on two distinct spine X-ray datasets (cervical and whole spine).
  • Achieved high performance metrics: Recall (96.82%, 94.38%), Precision (97.24%, 98.14%), Dice (97.03%, 96.22%), and IoU (94.24%, 92.72%).
  • Demonstrated superior performance compared to existing state-of-the-art vertebral segmentation techniques.

Conclusions:

  • The proposed deep learning and marker-controlled watershed method effectively segments vertebrae, accurately resolving adjacent vertebral adhesion.
  • This technique offers a significant advancement in automatic vertebral segmentation for spinal disease diagnosis.
  • The method shows strong potential for clinical adoption in medical imaging analysis.