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

Two methods for validating brain tissue classifiers.

Marcos Martin-Fernandez1, Sylvain Bouix, Lida Ungar

  • 1Laboratory of Mathematics in Imaging, Brigham and Women's Hospital, Boston, MA, USA.

Medical Image Computing and Computer-Assisted Intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention
|May 12, 2006
PubMed
Summary

This study evaluates seven automatic brain tissue classifiers using agreement measures. Williams

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

Plasma Phosphorylated Tau 217 in Participants at Risk for Chronic Traumatic Encephalopathy.

JAMA network open·2026
Same author

Shared and specific associations of amygdala nuclei volumes with PTSD symptom domains and childhood trauma: An ENIGMA-PGC PTSD mega-analysis.

Molecular psychiatry·2026
Same author

Youth Soccer Participation and Brain Health Outcomes in Adolescent Athletes.

JAMA network open·2026
Same author

Comparative profiles of anomalous self-experiences and social cognition in clinical high risk for psychosis and autism spectrum disorder.

Schizophrenia research. Cognition·2026
Same author

Brain regional susceptibility to tauopathy in individuals at risk for chronic traumatic encephalopathy.

Alzheimer's & dementia : the journal of the Alzheimer's Association·2026
Same author

Dose-dependent white matter changes associated with repetitive head impacts in former American football players.

Brain communications·2026

Area of Science:

  • Neuroimaging
  • Medical Image Analysis
  • Computational Neuroscience

Background:

  • Accurate brain tissue segmentation is crucial for neurological research and clinical diagnosis.
  • Evaluating the performance of automatic segmentation methods requires robust agreement measures.
  • Existing methods often rely on ground truth, which can be labor-intensive to obtain.

Purpose of the Study:

  • To evaluate and compare the performance of seven automatic brain tissue classifiers.
  • To assess the utility of various agreement measures for comparing segmentation techniques.
  • To introduce and validate a novel evaluation technique, Williams' index, for segmentation performance.

Main Methods:

  • Evaluation of seven automatic brain tissue classifiers using Simultaneous Truth and Performance Level Estimation (STAPLE).

Related Experiment Videos

  • Introduction and application of Williams' index as a novel evaluation metric.
  • Comparison of STAPLE and Williams' index on a dataset of 40 subjects with SPGR and T2-weighted MRI scans.

    • Main Results:

    • Demonstrated the comparability of STAPLE analysis and Williams' index in evaluating segmentation performance.
    • Showcased the practical application of agreement measures to compare different segmentation techniques.
    • Provided an interpretation of the results from the evaluated classifiers and metrics.

    Conclusions:

    • Williams' index offers a fast and easy-to-compute alternative for evaluating brain tissue segmentation when ground truth is unavailable.
    • The study provides valuable insights into the performance of different automatic brain tissue classifiers.
    • Agreement measures are effective tools for rigorous comparison of neuroimaging segmentation methods.