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

Normalized mutual information based registration using k-means clustering and shading correction.

Z F Knops1, J B A Maintz, M A Viergever

  • 1Utrecht University, Department of Computer Science, P.O. Box 80089, NL-3508 TB Utrecht, The Netherlands. zeger@cs.uu.nl

Medical Image Analysis
|August 23, 2005
PubMed
Summary
This summary is machine-generated.

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

Deep learning-based recognition of key anatomical structures during robot-assisted minimally invasive esophagectomy.

Surgical endoscopy·2023
Same author

Computer-aided anatomy recognition in intrathoracic and -abdominal surgery: a systematic review.

Surgical endoscopy·2022
Same author

Corrigendum to "Effects of hunger state on the brain responses to food cues across the life span" [NeuroImage 171 (2018) 246-255].

NeuroImage·2021
Same author

Fast contour propagation for MR-guided prostate radiotherapy using convolutional neural networks.

Medical physics·2019
Same author

Automatic extraction of the intracranial volume in fetal and neonatal MR scans using convolutional neural networks.

NeuroImage. Clinical·2019
Same author

Automatic brain tissue segmentation in fetal MRI using convolutional neural networks.

Magnetic resonance imaging·2019

This study enhances medical image registration using k-means intensity clustering and shading correction. These methods improve accuracy and robustness by reducing misregistrations in MR, CT, and PET scans.

Area of Science:

  • Medical imaging
  • Image processing
  • Computational anatomy

Background:

  • Mutual information (MI) based image registration is crucial for aligning medical images.
  • Standard registration methods can be sensitive to intensity variations and image artifacts like shading.
  • Equidistant re-binning, a common technique, may not optimally represent intensity distributions.

Purpose of the Study:

  • To investigate the impact of intensity clustering and shading correction on mutual information based image registration.
  • To improve the robustness and accuracy of medical image registration, particularly for MR images affected by inhomogeneities.
  • To compare a novel approach using k-means clustering and shading correction against standard registration techniques.

Main Methods:

  • Utilized k-means clustering for a more natural binning of image intensity distributions, replacing equidistant re-binning.

Related Experiment Videos

  • Implemented a shading correction method to mitigate the adverse effects of image inhomogeneities, common in MR scans.
  • Validated the proposed method on diverse datasets including clinical MR, CT, PET images, and synthetic MR data.

    • Main Results:

    • The combined approach of intensity clustering and shading correction significantly reduced the number of misregistrations.
    • The proposed method demonstrated increased robustness compared to standard registration without inhomogeneity correction and equidistant binning.
    • Accuracy was maintained while enhancing the overall reliability of the image registration process.

    Conclusions:

    • K-means intensity clustering and shading correction are effective enhancements for mutual information based image registration.
    • The developed method improves registration robustness and reduces errors, especially in the presence of image artifacts and intensity variations.
    • This approach offers a more reliable solution for aligning multimodal and unПартнер medical images.