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

Image restoration and reconstruction with a Bayesian approach

C M Kao1, X Pan, C T Chen

  • 1Department of Radiology, University of Chicago, Illinois 60637, USA. c-kao@uchicago.edu

Medical Physics
|June 3, 1998
PubMed
Summary

This study enhances Bayesian image restoration by adding diagonal line sites and hyperparameters, improving image quality and reducing artifacts. The improved method benefits positron emission tomography (PET) image reconstruction.

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

Quantum critical behavior of cuprate superconductors observed by inelastic X-ray scattering.

Nature communications·2026
Same author

Bimagnon dispersion of La<sub>2</sub>CuO<sub>4</sub> probed by resonant inelastic X-ray scattering.

Scientific reports·2025
Same author

Spin-valley coupling enhanced high-T<sub>C</sub> ferromagnetism in a non-van der Waals monolayer Cr<sub>2</sub>Se<sub>3</sub> on graphene.

Nature communications·2025
Same author

A model for decoding resistance in precision oncology: acquired resistance to FGFR inhibitors in cholangiocarcinoma.

Annals of oncology : official journal of the European Society for Medical Oncology·2024
Same author

Zhou et al. Reply.

Physical review letters·2024
Same author

Octahedral Distortion and Displacement-Type Ferroelectricity with Switchable Photovoltaic Effect in a 3d^{3}-Electron Perovskite System.

Physical review letters·2023

Area of Science:

  • Computer Vision
  • Image Processing
  • Bayesian Inference

Background:

  • Traditional Bayesian methods for image restoration and reconstruction can struggle with artifacts.
  • Existing neighborhood configurations may not fully capture image features like edges.

Purpose of the Study:

  • To extend Johnson's Bayesian method for enhanced image restoration and reconstruction.
  • To investigate the impact of novel features like diagonal line sites and hyperparameters on image quality.
  • To improve the performance of Bayesian image reconstruction, particularly for Positron Emission Tomography (PET) data.

Main Methods:

  • Introduction of diagonal line sites and symmetric neighborhood configurations into the Bayesian framework.
  • Development of a general formulation for arbitrary neighborhood configurations.

Related Experiment Videos

  • Estimation of a new hyperparameter specifically for line sites.
  • Extensive computer simulations to evaluate the effects of hyperparameters, diagonal sites, and neighborhood size.

    • Main Results:

    • Optimal performance requires distinct hyperparameters for intensity and line sites.
    • Larger neighborhood configurations generally yield better results.
    • Diagonal line sites and symmetric configurations effectively reduce blocky edge artifacts.
    • Significant quality improvement observed in restored images and PET reconstructions (simulated and real data).

    Conclusions:

    • The extended Bayesian method with diagonal line sites offers superior image restoration and reconstruction.
    • The findings provide a more robust approach for handling image artifacts and enhancing detail.
    • The method demonstrates practical utility in medical imaging, specifically for PET scans.