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

A fast fully 4-D incremental gradient reconstruction algorithm for list mode PET data.

Quanzheng Li1, Evren Asma, Sangtae Ahn

  • 1Signal and Image Processing Institute, University of Southern California, Los Angeles, CA 90089, USA.

IEEE Transactions on Medical Imaging
|January 25, 2007
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

Microelectrode arrays enable directional stereo-EEG during kainate-mediated seizures.

bioRxiv : the preprint server for biology·2026
Same author

An evaluation of brain volume and cortical thickness measurement at 0.55 T.

Magma (New York, N.Y.)·2026
Same author

Editorial: Deep learning in brain-computer interfaces.

Frontiers in human neuroscience·2026
Same author

Risk factors for psychological inflexibility among caregivers of children with hearing loss in China.

International journal of audiology·2026
Same author

Neural Responses to Affective Sentences Reveal Signatures of Depression.

Translational psychiatry·2026
Same author

Chronic Anemia Patients Demonstrate Diffuse Demyelination.

American journal of hematology·2026

A new four-dimensional incremental gradient (4DIG) algorithm offers faster, globally convergent estimation of continuous-time tracer density from positron emission tomography (PET) data. This method improves upon existing techniques for enhanced image reconstruction.

Area of Science:

  • Medical Imaging
  • Nuclear Medicine
  • Computational Science

Background:

  • Positron Emission Tomography (PET) imaging relies on accurate estimation of tracer distribution.
  • List mode data acquisition in PET presents unique challenges for image reconstruction.
  • Existing algorithms may face limitations in speed and convergence for continuous-time dynamic PET analysis.

Purpose of the Study:

  • To introduce a novel, fast, and globally convergent algorithm for continuous-time tracer density estimation.
  • To improve the accuracy and efficiency of image reconstruction from list mode PET data.
  • To provide a robust method for analyzing dynamic PET studies.

Main Methods:

  • Development of a fully four-dimensional incremental gradient (4DIG) algorithm.

Related Experiment Videos

  • Modeling photon pair detection as an inhomogeneous Poisson process.
  • Parameterization of rate functions using cubic B-splines and minimization of a cost function including log-likelihood and penalty terms.
  • Derivation of a computable bound for temporal basis function coefficients to ensure convergence.

    • Main Results:

    • The 4DIG algorithm demonstrates substantially faster convergence compared to preconditioned conjugate gradient methods in 4-D simulations.
    • The algorithm successfully estimates continuous-time tracer density from list mode PET data.
    • Four-dimensional reconstructions of real PET data validate the algorithm's performance.

    Conclusions:

    • The 4DIG algorithm provides a significant advancement in the speed and convergence of dynamic PET image reconstruction.
    • This method offers a more efficient approach to analyzing tracer kinetics in PET imaging.
    • The 4DIG algorithm shows promise for improved diagnostic capabilities in nuclear medicine.