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

Artificially enlarged training set in image segmentation.

Tuomas Tölli1, Juha Koikkalainen, Kirsi Lauerma

  • 1Laboratory of Biomedical Engineering, Helsinki University of Technology, P.O.B. 2200, FIN-02015 HUT, Finland. tuomas.tolli@tkk.fi

Medical Image Computing and Computer-Assisted Intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention
|March 16, 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

Orbitofrontal atrophy on MRI appears to be an indicator of C9orf72 repeat expansion status in FTD.

Journal of neuroradiology = Journal de neuroradiologie·2026
Same author

Brain ventricle morphology markers in predicting shunt surgery outcome in idiopathic normal-pressure hydrocephalus.

Fluids and barriers of the CNS·2026
Same author

Neighborhood-Level Disadvantage Impacts Multiple Measures of Brain Health: An Imaging Epidemiology Study.

bioRxiv : the preprint server for biology·2026
Same author

Dementia Care Research and Psychosocial Factors.

Alzheimer's & dementia : the journal of the Alzheimer's Association·2025
Same author

Biomarkers.

Alzheimer's & dementia : the journal of the Alzheimer's Association·2025
Same author

Clinical Manifestations.

Alzheimer's & dementia : the journal of the Alzheimer's Association·2025

To improve medical image segmentation, researchers artificially enlarged training datasets for statistical shape models. This method reduced segmentation errors, achieving an average error of 2.11 mm for cardiac models.

Area of Science:

  • Medical Imaging
  • Computer Vision
  • Biomedical Engineering

Background:

  • Statistical shape models (SSMs) are crucial for medical image segmentation.
  • Limited training data in SSMs often leads to excessive deformation constraints.
  • Artificial training set enlargement is a proposed solution to overcome data limitations.

Purpose of the Study:

  • To analyze error sources in SSM-based segmentation.
  • To improve optimization processes for SSM segmentation.
  • To evaluate an artificial training set enlargement method for 3D cardiac MR data.

Main Methods:

  • Analysis of error sources within SSM-based segmentation.
  • Improvement of optimization algorithms for SSMs.
  • Implementation of an artificial data enlargement technique using non-rigid transformations.

Related Experiment Videos

  • Evaluation on 3D cardiac Magnetic Resonance (MR) volume data.

    • Main Results:

    • The artificial enlargement method effectively addressed deformation constraints.
    • The improved optimization processes enhanced segmentation accuracy.
    • Evaluation on 25 subjects showed an average error of 2.11 mm for the four-chamber model with 250 artificial modes.

    Conclusions:

    • Artificial enlargement of training sets is a viable strategy to improve SSM segmentation performance.
    • The proposed method significantly reduces segmentation errors in 3D cardiac MR imaging.
    • This approach offers a practical solution for segmentation tasks with limited datasets.