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

Correction: Carbon and energy metabolism for the mixotrophic culture of <i>Chlorella vulgaris</i> using sodium acetate as a carbon source.

Frontiers in microbiology·2026
Same author

Hairyvein agrimony-derived nanovesicles in microneedles eradicate intracellular bacteria and reverse senescence to heal diabetic foot osteomyelitis.

Journal of controlled release : official journal of the Controlled Release Society·2026
Same author

Defect-heterointerface synergy in CuO/S-SnO<sub>2</sub> heterostructures for highly sensitive n-butanol detection.

Talanta·2026
Same author

Nature-derived exosome-like nanoparticles (NELNs) for intracellular transport and targeted RNA delivery.

Biotechnology advances·2026
Same author

Mrgpra2<sup>+</sup> neutrophils integrate infection-derived signals to trigger NET-mediated antimicrobial defense in bone marrow.

Cell reports·2026
Same author

Kinetic and Passivation Limitations in Titanium Reduction via Molten Oxide-Fluoride Electrolysis.

Chemphyschem : a European journal of chemical physics and physical chemistry·2026

Related Experiment Video

Updated: Oct 13, 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

3.0K

Automatic vertebrae localization and segmentation in CT with a two-stage Dense-U-Net.

Pengfei Cheng1, Yusheng Yang2, Huiqiang Yu2

  • 1Shanghai Key Laboratory of Intelligent Manufacturing and Robotics, Shanghai University, Shanghai, 200444, China. jonah0712@shu.edu.cn.

Scientific Reports
|November 13, 2021
PubMed
Summary

This study presents a deep learning method for automatically locating and segmenting vertebrae in CT scans. The approach achieves high accuracy in both localization and segmentation, demonstrating its effectiveness for spinal image analysis.

More Related Videos

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
14:08

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Published on: April 13, 2013

42.9K
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

558

Related Experiment Videos

Last Updated: Oct 13, 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

3.0K
Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
14:08

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Published on: April 13, 2013

42.9K
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

558

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Spinal Anatomy

Background:

  • Automatic vertebrae localization and segmentation in CT scans are crucial for spinal analysis and computer-assisted surgery.
  • High anatomical variation in patients presents significant challenges to these automated processes.

Purpose of the Study:

  • To develop and evaluate a deep learning approach for precise automatic vertebrae localization and segmentation in CT images.
  • To address the challenges posed by anatomical variations in the spine.

Main Methods:

  • A two-stage deep learning model, a Dense-U-Net architecture, was employed.
  • The first stage utilized a 2D-Dense-U-Net for vertebrae centroid detection and localization.
  • The second stage employed a 3D-Dense-U-Net for precise segmentation within identified regions of interest.

Main Results:

  • The method achieved excellent performance on the CSI 2014 dataset with a location error of 1.69 ± 0.78 mm and a 100% detection rate for localization.
  • Vertebrae segmentation yielded a dice coefficient of 0.953 ± 0.014 and intersection over union of 0.911 ± 0.025.
  • The approach demonstrated generalizability on the xVertSeg challenge dataset, showing a dice coefficient of 0.877 ± 0.035.

Conclusions:

  • The proposed two-stage deep learning method effectively automates vertebrae localization and segmentation in CT scans.
  • The approach proves efficient and generalizable, offering a robust solution for spinal image analysis and surgical applications.