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

Bayesian approach to segmentation of statistical parametric maps.

J C Rajapakse1, J Piyaratna

  • 1School of Computer Engineering, Nanyang Technological University, Singapore, Singapore. asjagath@ntu.edu.sg

IEEE Transactions on Bio-Medical Engineering
|October 5, 2001
PubMed
Summary

This study introduces a novel contextual segmentation technique using a Markov random field (MRF) for detecting brain activation in functional brain images. The MRF approach offers superior segmentation compared to existing methods, enhancing the analysis of brain activity.

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

Two-stage multi-class support vector machines to protein secondary structure prediction.

Pacific Symposium on Biocomputing. Pacific Symposium on Biocomputing·2005
Same author

Proton spectroscopic imaging shows abnormalities in glial and neuronal cell pools in frontal lobe epilepsy.

Epilepsia·2002
Same author

Probabilitic fusion of hemodynamic parameter maps.

Critical reviews in biomedical engineering·2000
Same author

Efficient sampling in 3-D stereologic volumetry.

Critical reviews in biomedical engineering·2000
Same author

Random-grid stereologic volumetry of MR head scans.

Journal of magnetic resonance imaging : JMRI·2000
Same author

Development of the human corpus callosum during childhood and adolescence: a longitudinal MRI study.

Progress in neuro-psychopharmacology & biological psychiatry·1999

Area of Science:

  • Neuroimaging
  • Computational Neuroscience
  • Medical Image Analysis

Background:

  • Functional brain imaging analysis often relies on statistical methods to detect activation.
  • Previous segmentation techniques have limitations in capturing contextual information of activated brain regions.
  • Statistical Parametric Mapping (SPM) is a common but context-free approach.

Purpose of the Study:

  • To develop and evaluate a novel contextual segmentation technique for detecting brain activation.
  • To improve upon existing methods by incorporating spatial context using a Markov random field (MRF).
  • To compare the proposed technique against simple thresholding and SPM approaches.

Main Methods:

  • A Bayesian framework incorporating a Markov random field (MRF) for contextual segmentation.

Related Experiment Videos

  • Utilizing likelihoods from statistical parametric maps (SPMs) for Maximum A Posteriori (MAP) estimation.
  • An iterative segmentation algorithm based on simulated annealing, capable of analyzing multi-stimuli experiments.

    • Main Results:

    • The MRF model effectively represents activation patterns in functional brain images.
    • The proposed contextual segmentation technique outperformed simple thresholding and SPM in simulations and real fMRI data.
    • Demonstrated superior segmentation accuracy for memory retrieval and working memory tasks.

    Conclusions:

    • Markov random fields provide a valid and effective representation for brain activation patterns.
    • The developed contextual segmentation technique offers a superior alternative to context-free and SPM approaches.
    • This method enhances the detection and analysis of brain activation in functional neuroimaging studies.