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 Videos

Active shape model segmentation with optimal features.

Bram van Ginneken1, Alejandro F Frangi, Joes J Staal

  • 1Image Sciences Institute, University Medical Center Utrecht, Heidelberglaan 100,3584 CX Utrecht, The Netherlands. bram@isi.uu.nl

IEEE Transactions on Medical Imaging
|December 11, 2002
PubMed
Summary
This summary is machine-generated.

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

Artificial intelligence for detecting fetal orofacial clefts and advancing medical education.

Nature communications·2026
Same author

Multi-omic analysis of deep learning-derived phenotypes links ophthalmic imaging to cardiovascular and neurological traits.

Nature cardiovascular research·2026
Same author

From pixels to polygons: A survey of deep learning approaches for medical image-to-mesh reconstruction.

Medical image analysis·2026
Same author

Fourier-Net+: Band-Limited Spatial Representation for Efficient Medical Image Registration.

IEEE transactions on neural networks and learning systems·2026
Same author

Angiography-free diagnosis of retinal diseases via interpretable multi-modal learning.

NPJ digital medicine·2026
Same author

An effective deep learning algorithm for medical image registration.

PLOS digital health·2026

This study introduces an improved active shape model for medical image segmentation. The enhanced method uses optimal local features and a nonlinear classifier, achieving significantly better segmentation accuracy in lung and brain imaging tasks.

Area of Science:

  • Medical image analysis
  • Computer vision
  • Computational anatomy

Background:

  • Active Shape Models (ASMs) are widely used for image segmentation.
  • The original ASM formulation relies on normalized first-order derivative profiles and linear classifiers.
  • Limitations exist in the original ASM's feature selection and classification approach.

Purpose of the Study:

  • To develop an improved active shape model segmentation scheme.
  • To enhance segmentation accuracy by utilizing optimal local features and a nonlinear classifier.
  • To automate feature selection for improved landmark displacement prediction.

Main Methods:

  • Implemented a novel ASM segmentation approach using optimal local features.
  • Employed a nonlinear kNN-classifier instead of a linear Mahalanobis distance for landmark displacement.

Related Experiment Videos

  • Utilized automatic feature selection via sequential forward and backward selection on training data.
  • Validated the method on synthetic data and real medical images (chest radiographs, MRI brain scans).

    • Main Results:

    • The new method demonstrated significantly improved segmentation results compared to the original ASM.
    • Overlap error was significantly reduced (p < 0.001) across all tested medical segmentation tasks.
    • Successful segmentation of lung fields, cerebellum, and corpus callosum was achieved.

    Conclusions:

    • The proposed ASM segmentation scheme offers superior performance over the original formulation.
    • Optimal local features and nonlinear classification enhance segmentation accuracy in medical imaging.
    • The automated feature selection process contributes to the robustness of the method.