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

Shape regression machine.

Shaohua Kevin Zhou1, Dorin Comaniciu

  • 1Integrated Data Systems Department,Siemens Corporate Research 755 College Road East, Princeton, NJ 08540, USA.

Information Processing in Medical Imaging : Proceedings of the ... Conference
|July 19, 2007
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

DuoMod-Net: Logarithmic balancing and geometric refinement for imbalanced semi-supervised medical image segmentation.

Patterns (New York, N.Y.)·2026
Same author

Multi-plane vision transformer for hemorrhage classification using axial and sagittal MRI data.

Scientific reports·2026
Same author

Deep learning based 3D brain metastasis synthesis with configurable parameters for 3D data augmentation.

Scientific reports·2026
Same author

A self-supervised framework for emphysema anomaly detection and staging in computed tomography scans.

Patterns (New York, N.Y.)·2026
Same author

Single shot full plan deep learning dose computation for radiation therapy using spherical harmonics.

Medical physics·2026
Same author

IRESeek: structure-informed deep learning method for accurate identification of internal ribosome entry sites in circular RNAs.

NAR genomics and bioinformatics·2026
Same journal

Enhancing Alzheimer's Diagnosis: Leveraging Anatomical Landmarks in Graph Convolutional Neural Networks on Tetrahedral Meshes.

Information processing in medical imaging : proceedings of the ... conference·2026
Same journal

Cycle-Consistent Zero-Shot Through-Plane Super-Resolution for Anisotropic Head MRI.

Information processing in medical imaging : proceedings of the ... conference·2026
Same journal

Brightness-Invariant Tracking Estimation in Tagged MRI.

Information processing in medical imaging : proceedings of the ... conference·2025
Same journal

Multi-View and Multi-Scale Alignment for Contrastive Language-Image Pre-training in Mammography.

Information processing in medical imaging : proceedings of the ... conference·2025
Same journal

Using Multiple Instance Learning to Build Multimodal Representations.

Information processing in medical imaging : proceedings of the ... conference·2025
Same journal

mSPD-NN: A Geometrically Aware Neural Framework for Biomarker Discovery from Functional Connectomics Manifolds.

Information processing in medical imaging : proceedings of the ... conference·2024
See all related articles

We developed a novel machine learning method, shape regression machine (SRM), for real-time medical image segmentation of deformable anatomic structures. SRM overcomes limitations of traditional methods by using context and an annotated database for accurate, initialization-free segmentation.

Area of Science:

  • Medical image analysis
  • Machine learning
  • Computational anatomy

Background:

  • Traditional shape segmentation methods (e.g., deformable models, Mumford-Shah, active appearance models) rely on restrictive assumptions and require good initialization.
  • These limitations hinder accurate segmentation of complex, deformable anatomic structures in medical imaging.

Purpose of the Study:

  • To introduce a novel machine learning approach, the shape regression machine (SRM), for real-time segmentation of deformable anatomic structures in medical images.
  • To overcome the limitations of traditional segmentation techniques by removing restrictive assumptions and initialization requirements.

Main Methods:

  • SRM is a two-stage approach leveraging medical context and an annotated database.
  • Stage 1: Object detection for initialization and a regression solution using a single scan.

Related Experiment Videos

  • Stage 2: Learning a nonlinear regressor for nonrigid shape prediction from image appearance, enhanced by a boosting regression approach for real-time performance.

    • Main Results:

    • SRM effectively segments deformable anatomic structures without traditional method restrictions.
    • The approach demonstrated successful segmentation of the left ventricle endocardium in echocardiograms.
    • A boosting regression variant enables real-time segmentation capabilities.

    Conclusions:

    • SRM offers a flexible and robust alternative for medical image segmentation of deformable shapes.
    • The method's ability to learn from context and appearance variations enhances segmentation accuracy.
    • SRM shows significant potential for real-time clinical applications, particularly in cardiac imaging.