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

SVM for density estimation and application to medical image segmentation.

Zhao Zhang1, Su Zhang, Chen-xi Zhang

  • 1Biomedical Instrument Institute, Shanghai Jiao Tong University, Shanghai 200030, China. z_ball@sjtu.edu.cn

Journal of Zhejiang University. Science. B
|April 15, 2006
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

A Dynamic Contour Evolution Algorithm for Cell Segmentation and Synaptic Tracking under Occlusion.

Chemical & biomedical imaging·2026
Same author

Photoredox-catalyzed synthesis of trifluoromethylated benzimidazo-fused heterocycles.

Organic & biomolecular chemistry·2026
Same author

[Moxibustion enhances the antitumor efficacy of Trametinib in breast cancer-bearing mice via MEK/ERK signaling pathway].

Zhen ci yan jiu = Acupuncture research·2026
Same author

Yellow Colorectal Mucosa: Chronic Rectal Schistosomiasis.

Digestive diseases and sciences·2026
Same author

Potential value of serum IL-6 and IL-8 as biomarkers for screening and severity stratification of diabetic retinopathy.

Clinica chimica acta; international journal of clinical chemistry·2026
Same author

Exploring the therapeutic mechanism of kaempferol in acute lung injury: Network pharmacology and experimental validation.

Tissue & cell·2026
Same journal

Halotolerant plant growth-promoting bacteria <i>Enterobacter</i> sp. Av16 and <i>Acinetobacter</i> sp. Av23 enhance seed germination and seedling photosynthesis of <i>Apocynum pictum</i> under salt stress.

Journal of Zhejiang University. Science. B·2026
Same journal

Integrative analytical and statistical framework for optimization of multiplex qPCR detection of <i>TGFBI</i> mutations in refractive surgery candidates.

Journal of Zhejiang University. Science. B·2026
Same journal

Regenerative potential of Schneiderian membrane-derived mesenchymal stem cells in sinus floor elevation model and calvarial defect model.

Journal of Zhejiang University. Science. B·2026
Same journal

Esculetin attenuates migraine-like pain via CGRP suppression and meningeal mast cell modulation in rat models.

Journal of Zhejiang University. Science. B·2026
Same journal

Cell proliferation and differentiation during epidermis renewal.

Journal of Zhejiang University. Science. B·2026
Same journal

HENMT1: an RNA methyltransferase in biology and disease.

Journal of Zhejiang University. Science. B·2026
See all related articles

This study introduces a novel medical image segmentation method using support vector machines (SVM) for density estimation. The approach integrates shape information, improving segmentation accuracy on various image types.

Area of Science:

  • Medical Imaging
  • Computer Vision
  • Machine Learning

Background:

  • Accurate medical image segmentation is crucial for diagnosis and treatment planning.
  • Traditional methods often struggle to incorporate structural shape information effectively.
  • Support Vector Machines (SVM) offer robust density estimation capabilities.

Purpose of the Study:

  • To develop an advanced medical image segmentation method.
  • To leverage SVM for density estimation to build a prior model of image features.
  • To enhance segmentation by incorporating shape information into the narrow level set method.

Main Methods:

  • Utilized Support Vector Machine (SVM) for density estimation to create a prior model.
  • Constructed a prior model based on image intensity and curvature profiles from training data.

Related Experiment Videos

  • Employed the narrow level set method, defining surface evolution via the prior model, not energy minimization.

    • Main Results:

    • Demonstrated successful segmentation on synthetic, MR, and ultrasonic images.
    • The SVM-based prior model consistently improved segmentation accuracy.
    • The method effectively incorporated object shape information into the segmentation process.

    Conclusions:

    • The proposed SVM-based medical image segmentation method is effective and accurate.
    • Integrating shape information via a prior model enhances segmentation compared to traditional level set methods.
    • This approach offers a consistent and sparse solution for complex image segmentation tasks.