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

Multi-level adaptive segmentation of multi-parameter MR brain images.

A Zavaljevski1, A P Dhawan, M Gaskil

  • 1System Engineering Group, GE Medical Systems, Milwaukee, WI, USA.

Computerized Medical Imaging and Graphics : the Official Journal of the Computerized Medical Imaging Society
|April 18, 2000
PubMed
Summary

This study introduces a new method for automatic brain image segmentation, classifying 15 tissue types from multi-parameter MR scans. The advanced model accurately segments brain structures, aiding in the diagnosis and monitoring of conditions like stroke.

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

Tabular foundation model predicts alternative lengthening of telomeres (ALT) and identifies SMARCAL1 as a target in ALT-driven cancers.

bioRxiv : the preprint server for biology·2026
Same author

Laser spectroscopy and CP-violation sensitivity of actinium monofluoride.

Nature·2025
Same author

Production and purification of molecular <sup>225</sup>Ac at CERN-ISOLDE.

Journal of radioanalytical and nuclear chemistry·2025
Same author

Production study of Fr, Ra and Ac radioactive ion beams at ISOLDE, CERN.

Scientific reports·2024
Same author

Electronic Health Records and Pulmonary Function Data: Developing an Interoperability Roadmap. An Official American Thoracic Society Workshop Report.

Annals of the American Thoracic Society·2021
Same author

The response of the ozone layer to quadrupled CO<sub>2</sub> concentrations.

Journal of climate·2020

Area of Science:

  • Medical Imaging
  • Computational Neuroscience
  • Radiology

Background:

  • Manual MR brain image segmentation is operator-dependent and difficult to reproduce.
  • Existing automatic methods typically classify limited tissue classes (e.g., white matter, gray matter, CSF).
  • Accurate segmentation is crucial for visualizing and quantifying brain structures, especially in patients with cerebrovascular deficiency (CVD) and stroke.

Purpose of the Study:

  • To present a novel model-based method for automatic segmentation and classification of multi-parameter MR brain images.
  • To develop a model capable of segmenting a larger number of clinically relevant brain tissue classes (15) for neuroradiologists.
  • To improve the accuracy and reproducibility of brain image analysis for conditions like stroke.

Main Methods:

Related Experiment Videos

  • A model-based approach using a Gauss Markov random field to approximate spatial distribution of 15 brain tissue classes.
  • Maximum likelihood estimation for class and transitional probabilities per pixel.
  • Utilized multi-parameter MR images (T(1), T(2), proton density, Gd+T(1), perfusion) and elastic transformation for image registration.
  • Ground truth established through manual pixel-by-pixel segmentation by neuroradiologists using a specialized interface.

Main Results:

  • The novel method achieved accurate segmentation and classification of multi-parameter MR brain images.
  • Results demonstrated efficacy and accuracy comparable to manual segmentation.
  • The model showed capability in creating and learning new tissue classes, adapting to individual patient data.

Conclusions:

  • The presented model-based method offers an effective and accurate solution for automatic multi-parameter MR brain image segmentation.
  • This approach enhances the ability to analyze brain structures for conditions like stroke and CVD.
  • The method's adaptability and capacity for learning new classes represent a significant advancement in clinical neuroimaging analysis.