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

DiffGeo-AOR: Diffusion-Optimized Medical Grading via Geometric Priors enhanced Autoregressive Ordinal Regression.

IEEE transactions on medical imaging·2026
Same author

Exploring the use of AI-generated counterfactual chest X-rays to enhance diagnostic learning in medical education.

BMC medical education·2026
Same author

Causal counterfactual simulation for treatment decisions in multimodal lung disease data.

Computers in biology and medicine·2026
Same author

Investigating the Effects of the UniReliever® Knee Brace Using Analysis of a Novel Digital Platform, MAI Motion.

Cureus·2026
Same author

Metabolic dysfunction over a life course key to healthy ageing inequality.

Aging clinical and experimental research·2025
Same author

Clinical features, myocardial injury and systolic impairment in acute myocarditis.

Open heart·2024

Related Experiment Video

Updated: Jun 7, 2026

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

Automatic Graph Cut Segmentation of Lesions in CT Using Mean Shift Superpixels.

Xujiong Ye1, Gareth Beddoe, Greg Slabaugh

  • 1R&D Department, Medicsight PLC, 66 Hammersmith Road, London W14 8UD, UK.

International Journal of Biomedical Imaging
|November 6, 2010
PubMed
Summary

This study introduces an automatic CT lesion extraction method using 5D feature vectors, mean shift clustering, and graph cuts. The approach accurately segments lung nodules and colonic polyps, improving lesion detection.

More Related Videos

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

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
10:25

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

Published on: September 25, 2019

Related Experiment Videos

Last Updated: Jun 7, 2026

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

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

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
10:25

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

Published on: September 25, 2019

Area of Science:

  • Medical Imaging
  • Computer Vision
  • Computational Pathology

Background:

  • Accurate lesion segmentation in CT data is crucial for diagnosis and treatment planning.
  • Existing methods often struggle with lesions adjacent to similar-intensity structures or those affected by partial volume effects.
  • Lesion segmentation requires robust methods that integrate multiple feature types.

Purpose of the Study:

  • To develop and evaluate a novel, automatic method for precise lesion extraction from CT images.
  • To leverage a combination of intensity, shape, and spatial features for improved segmentation accuracy.
  • To establish an automated seeding mechanism for graph cut segmentation based on lesion morphology.

Main Methods:

  • A five-dimensional (5D) feature vector (intensity, shape index, 3D spatial location) is computed for each voxel.
  • Nonparametric mean shift clustering is employed to form superpixels based on the 5D features, enabling oversegmentation.
  • A graph cut algorithm refines segmentation by grouping superpixels using a novel energy formulation incorporating shape, intensity, and spatial information.
  • Automatic foreground and background seeds are derived from detected spherical subregions within lesions.

Main Results:

  • The method successfully segmented various lesions, including lung nodules and colonic polyps, in a clinical CT dataset.
  • Quantitative evaluation on 101 solid and 80 GGO nodules demonstrated the method's potential.
  • The integration of spatial-intensity-shape features proved effective for segmenting challenging lesions, such as those near similar-intensity structures or exhibiting partial volume effects.
  • Mean shift superpixels enhanced result robustness and reduced computation time.

Conclusions:

  • The proposed automatic method offers accurate and robust lesion extraction from CT data.
  • The combined use of mean shift superpixels and graph cuts with a novel energy formulation provides a powerful approach for medical image segmentation.
  • This technique shows significant potential for improving the detection and analysis of lesions in clinical practice.