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

Choosing parameters in block-iterative or ordered subset reconstruction algorithms.

Charles Byrne1

  • 1Department of Mathematical Sciences, University of Massachusetts Lowell, Lowell, MA 01854, USA. Charles_Byrne@uml.edu

IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society
|March 15, 2005
PubMed
Summary

This study clarifies how relaxation and normalization parameters accelerate convergence in iterative algorithms for linear systems, crucial for applications like emission tomography. Proper parameter selection is key to improving computational efficiency.

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

Quantifying the spatial risk of Avian Influenza introduction into British poultry by wild birds.

Scientific reports·2019
Same author

On the Use of Tartar-Emetic Ointment in Chorea, &c.

The London medical and physical journal·2018
Same author

Childhood trauma levels in individuals attending adult mental health services: An evaluation of clinical records and structured measurement of childhood trauma.

Child abuse & neglect·2015
Same author

Improving the uptake of systematic reviews: a systematic review of intervention effectiveness and relevance.

BMJ open·2014
Same author

Making evidence more wanted: a systematic review of facilitators to enhance the uptake of evidence from systematic reviews and meta-analyses.

International journal of evidence-based healthcare·2012

Area of Science:

  • Applied Mathematics
  • Image Reconstruction
  • Computational Science

Background:

  • Iterative algorithms for solving linear systems (y=Px) are vital in fields like emission tomography.
  • Standard methods such as expectation maximization maximum likelihood and simultaneous multiplicative algebraic reconstruction technique exhibit slow convergence on large datasets.
  • Block-iterative and ordered-subset methods offer potential acceleration but introduce complex parameters.

Purpose of the Study:

  • To investigate the theoretical foundations of block-iterative algorithms for linear systems.
  • To elucidate the critical roles of relaxation and normalization parameters in accelerating convergence.
  • To provide guidance for optimal parameter selection in iterative reconstruction techniques.

Main Methods:

Related Experiment Videos

  • Analysis of linear systems of equations (y=Px) with nonnegative matrices and positive vectors.
  • Examination of block-iterative and ordered-subset acceleration strategies.
  • Theoretical discussion of parameter influence on convergence rates.

Main Results:

  • Block-iterative acceleration is not inherently faster; parameter choice is crucial.
  • Understanding the theoretical underpinnings reveals the precise function of relaxation and normalization parameters.
  • Correct parameter selection significantly impacts the convergence speed of iterative algorithms.

Conclusions:

  • A deeper theoretical understanding of iterative algorithms and their parameters is essential for efficient image reconstruction.
  • Optimizing parameter selection in block-iterative methods leads to significant improvements in computational performance.
  • This work provides foundational insights for users of expectation maximization maximum likelihood and simultaneous multiplicative algebraic reconstruction technique variants.