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

Implementation of a model-based nonuniform scatter correction scheme for SPECT

A Welch1, G T Gullberg

  • 1Department of Radiology, The University of Utah, Salt Lake City 84112, USA. a.welch@abdn.ac.uk

IEEE Transactions on Medical Imaging
|April 9, 1998
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

Regional and hemispheric susceptibility of the temporal lobe to FTLD-TDP type C pathology.

NeuroImage. Clinical·2020
Same author

Molecular strategies used by hibernators: Potential therapeutic directions for ischemia reperfusion injury and preservation of human donor organs.

Transplantation reviews (Orlando, Fla.)·2019
Same author

Exploring the association between ingestion of foods with higher potential salicylate content and symptom exacerbation in chronic rhinosinusitis. Data from the National Chronic Rhinosinusitis Epidemiology Study.

Rhinology·2019
Same author

Predictors for routine admission to paediatric intensive care for post-supraglottoplasty laryngomalacia patients.

The Journal of laryngology and otology·2017
Same author

Tumour expression of leptin is associated with chemotherapy resistance and therapy-independent prognosis in gastro-oesophageal adenocarcinomas.

British journal of cancer·2016
Same author

Tumour expression of leptin is associated with chemotherapy resistance and therapy-independent prognosis in gastro-oesophageal adenocarcinomas.

British journal of cancer·2015

Four scatter-compensation schemes were evaluated for image reconstruction using the expectation-maximization maximum-likelihood (EM-ML) algorithm. Faster subtraction/addition schemes offer practical clinical utility, while incorporating the model in both forward and back projectors yields superior image quality.

Area of Science:

  • Medical Imaging
  • Image Reconstruction
  • Computational Imaging

Background:

  • Scatter in imaging systems degrades image quality and quantification accuracy.
  • Accurate scatter compensation is crucial for reliable image analysis.
  • Previous work established a two-dimensional (2-D) scatter model for compensation.

Purpose of the Study:

  • To evaluate four novel scatter-compensation schemes based on a 2-D scatter model.
  • To compare the accuracy, convergence, and noise properties of these schemes.
  • To determine the most suitable scheme for clinical applications.

Main Methods:

  • Iterative expectation-maximization maximum-likelihood (EM-ML) algorithm used for reconstruction.
  • Four schemes tested: model in forward/back projector, forward projector only, subtraction, and addition.

Related Experiment Videos

  • Monte Carlo simulated projection data from uniform and non-uniform objects used for evaluation.

    • Main Results:

    • All four schemes produced images with accuracy comparable to scatter-free data.
    • Subtraction scheme demonstrated faster convergence in iterations and time.
    • Forward projector and scatter-addition schemes showed slightly better signal-to-noise ratio (SNR) than subtraction.
    • Model in both forward and back projectors yielded the highest SNR, surpassing perfect scatter rejection.

    Conclusions:

    • All evaluated scatter-compensation schemes effectively improve image accuracy.
    • Faster subtraction/addition schemes are promising for routine clinical use due to efficiency.
    • The scheme with the model in both forward and back projectors offers the best noise performance but is computationally intensive.