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 Video

Updated: Jul 3, 2026

Automated Joint Space Detection Improves Bone Segmentation Accuracy
06:45

Automated Joint Space Detection Improves Bone Segmentation Accuracy

Published on: November 28, 2025

Effects of registration regularization and atlas sharpness on segmentation accuracy.

B T Thomas Yeo1, Mert R Sabuncu, Rahul Desikan

  • 1Computer Science and Artificial Intelligence Laboratory, Massachusetts Institute of Technology, Cambridge, MA, USA. ythomas@csail.mit.edu

Medical Image Analysis
|August 1, 2008
PubMed
Summary

This study introduces a generative model for joint image registration and segmentation. It finds an optimal balance between atlas sharpness and warp regularization for improved brain parcellation, outperforming existing methods.

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

Excessive Censoring Degrades Individual-Specific Cortical Parcellations and Personalized TMS Targets.

bioRxiv : the preprint server for biology·2026
Same author

Atrophy in preclinical Alzheimer's disease maps to a network that predicts longitudinal decline.

Molecular psychiatry·2026
Same author

Individual-specific resting-state networks predict language dominance in drug-resistant epilepsy.

Epilepsia·2026
Same author

Widespread use of invalid statistical tests in biomedical machine learning.

bioRxiv : the preprint server for biology·2026
Same author

Developing a multi-modal neuroimaging-based BrainAge model across childhood.

bioRxiv : the preprint server for biology·2026
Same author

Convergent and divergent brain-cognition development in early adolescence.

Nature communications·2026

Area of Science:

  • Medical Image Analysis
  • Computational Neuroscience
  • Machine Learning

Background:

  • Non-rigid registration and atlas-based segmentation often involve empirical tradeoffs between regularization and image fidelity.
  • This empirical approach results in probabilistic atlases with variable sharpness, impacting segmentation accuracy.
  • Existing methods lack a principled framework for optimizing this tradeoff.

Purpose of the Study:

  • To develop a generative model for joint image registration and segmentation.
  • To establish a principled method for constructing unbiased atlases at optimal sharpness.
  • To investigate the impact of atlas sharpness and warp smoothness on cortical surface parcellation.

Main Methods:

  • Employed a generative model for joint registration and segmentation.

More Related Videos

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
10:25

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

Published on: September 25, 2019

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
14:08

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Published on: April 13, 2013

Related Experiment Videos

Last Updated: Jul 3, 2026

Automated Joint Space Detection Improves Bone Segmentation Accuracy
06:45

Automated Joint Space Detection Improves Bone Segmentation Accuracy

Published on: November 28, 2025

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
10:25

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

Published on: September 25, 2019

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
14:08

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Published on: April 13, 2013

  • Atlas construction as parameter estimation within the generative framework.
  • Studied the tradeoff between atlas sharpness and warp smoothness in cortical surface parcellation.

    • Main Results:

    • The generative model allows computation of unbiased atlases at various sharpness levels.
    • Optimal segmentation (parcellation) achieved through a unique balance of atlas sharpness and warp regularization.
    • Statistically significant improvements in parcellation accuracy compared to the FreeSurfer algorithm.
    • Demonstrated that a single, optimally sharp atlas can be used for registration-segmentation with fixed warp constraints.

    Conclusions:

    • A generative framework provides an optimal balance for atlas sharpness and warp regularization in image analysis.
    • This approach yields statistically significant improvements in brain parcellation accuracy.
    • Optimal atlas sharpness and warp smoothness can be determined empirically from training data.
    • Segmentation accuracy shows tolerance to minor mismatches between atlas sharpness and warp smoothness.