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: Jun 20, 2026

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

Bayesian registration via local image regions: information, selection and marginalization.

Matthew Toews1, William M Wells

  • 1Brigham and Women's Hospital, Harvard Medical School, USA. mt@bwh.harvard.edu

Information Processing in Medical Imaging : Proceedings of the ... Conference
|August 22, 2009
PubMed
Summary

This study introduces a new Bayesian approach for robust image registration, improving accuracy by considering location as a variable. This method enhances medical image alignment for applications like image-guided neurosurgery.

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

Equivariant Filters for Efficient Tracking in 3D Imaging.

Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention·2026
Same author

Sample-Specific Debiasing for Better Image-Text Models.

Proceedings of machine learning research·2026
Same author

Intraoperative Registration by Cross-Modal Inverse Neural Rendering.

Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention·2025
Same author

Two Projections Suffice for Cerebral Vascular Reconstruction.

Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention·2025
Same author

Representative scale-invariant characteristics of male and female brains in magnetic resonance images.

Neuroimage. Reports·2025
Same authorSame journal

Using Multiple Instance Learning to Build Multimodal Representations.

Information processing in medical imaging : proceedings of the ... conference·2025

Area of Science:

  • Medical image analysis
  • Computational imaging
  • Neurosurgery

Background:

  • Accurate image registration is crucial for image-guided neurosurgery.
  • Existing methods like global or iconic registration have limitations in robustness and accuracy.

Purpose of the Study:

  • To develop a novel Bayesian registration formulation for improved robustness.
  • To establish a link between Bayesian registration and mutual information (MI).
  • To introduce an MI-based technique for selecting informative image regions.

Main Methods:

  • Proposed a Bayesian registration formulation treating image location as a latent random variable.
  • Marginalized location to determine the maximum a priori (MAP) transform.
  • Established a mathematical link between the Bayesian formulation and MI.

More Related Videos

Human Brown Adipose Tissue Depots Automatically Segmented by Positron Emission Tomography/Computed Tomography and Registered Magnetic Resonance Images
09:21

Human Brown Adipose Tissue Depots Automatically Segmented by Positron Emission Tomography/Computed Tomography and Registered Magnetic Resonance Images

Published on: February 18, 2015

Related Experiment Videos

Last Updated: Jun 20, 2026

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

Human Brown Adipose Tissue Depots Automatically Segmented by Positron Emission Tomography/Computed Tomography and Registered Magnetic Resonance Images
09:21

Human Brown Adipose Tissue Depots Automatically Segmented by Positron Emission Tomography/Computed Tomography and Registered Magnetic Resonance Images

Published on: February 18, 2015

  • Developed an MI-based region selection technique using image intensity and spatial location.

    • Main Results:

    • The marginalization formulation provides more robust registration than global or iconic methods.
    • The MI-based region selection effectively identifies informative areas for registration.
    • Demonstrated successful ultrasound (US) to magnetic resonance (MR) registration in a neurosurgical context.

    Conclusions:

    • The proposed Bayesian marginalization formulation enhances image registration robustness.
    • MI-based region selection is an effective strategy for improving registration accuracy.
    • The developed techniques show significant promise for image-guided neurosurgical applications.