Fast gradient-based methods for Bayesian reconstruction of transmission and emission PET images

Erkan U. Mumcuoglu, Richard Leahy, Simon R. Cherry, Simon R Cherry

Research output: Contribution to journalArticle

187 Citations (Scopus)

Abstract

We describe conjugate gradient algorithms for reconstruction of transmission and emission PET images. The reconstructions are based on a Bayesian formulation, where the data are modeled as a collection of independent Poisson random variables and the image is modeled using a Markov random field. A conjugate gradient algorithm is used to compute a maximum a posteriori (MAP) estimate of the image by maximizing over the posterior density. To ensure nonnegativity of the solution, a penalty function is used to convert the problem to one of unconstrained optimization. Preconditioners are used to enhance convergence rates. These methods generally achieve effective convergence in 15-25 iterations. Reconstructions are presented of an 18FDG whole body scan from data collected using a Siemens/CTI ECAT931 whole body system. These results indicate significant improvements in emission image quality using the Bayesian approach, in comparison to filtered backprojection, particularly when reprojections of the MAP transmission image are used in place of the standard attenuation correction factors.

Original languageEnglish (US)
Pages (from-to)687-701
Number of pages15
JournalIEEE Transactions on Medical Imaging
Volume13
Issue number4
DOIs
StatePublished - Dec 1994
Externally publishedYes

Fingerprint

Bayes Theorem
Whole Body Imaging
Image communication systems
Fluorodeoxyglucose F18
Random variables
Image quality

ASJC Scopus subject areas

  • Computational Theory and Mathematics
  • Computer Science Applications
  • Biomedical Engineering
  • Electrical and Electronic Engineering
  • Radiology Nuclear Medicine and imaging
  • Radiological and Ultrasound Technology

Cite this

Fast gradient-based methods for Bayesian reconstruction of transmission and emission PET images. / Mumcuoglu, Erkan U.; Leahy, Richard; Cherry, Simon R.; Cherry, Simon R.

In: IEEE Transactions on Medical Imaging, Vol. 13, No. 4, 12.1994, p. 687-701.

Research output: Contribution to journalArticle

Mumcuoglu, Erkan U. ; Leahy, Richard ; Cherry, Simon R. ; Cherry, Simon R. / Fast gradient-based methods for Bayesian reconstruction of transmission and emission PET images. In: IEEE Transactions on Medical Imaging. 1994 ; Vol. 13, No. 4. pp. 687-701.
@article{e288673dde744c54a29d56da997d7217,
title = "Fast gradient-based methods for Bayesian reconstruction of transmission and emission PET images",
abstract = "We describe conjugate gradient algorithms for reconstruction of transmission and emission PET images. The reconstructions are based on a Bayesian formulation, where the data are modeled as a collection of independent Poisson random variables and the image is modeled using a Markov random field. A conjugate gradient algorithm is used to compute a maximum a posteriori (MAP) estimate of the image by maximizing over the posterior density. To ensure nonnegativity of the solution, a penalty function is used to convert the problem to one of unconstrained optimization. Preconditioners are used to enhance convergence rates. These methods generally achieve effective convergence in 15-25 iterations. Reconstructions are presented of an 18FDG whole body scan from data collected using a Siemens/CTI ECAT931 whole body system. These results indicate significant improvements in emission image quality using the Bayesian approach, in comparison to filtered backprojection, particularly when reprojections of the MAP transmission image are used in place of the standard attenuation correction factors.",
author = "Mumcuoglu, {Erkan U.} and Richard Leahy and Cherry, {Simon R.} and Cherry, {Simon R}",
year = "1994",
month = "12",
doi = "10.1109/42.363099",
language = "English (US)",
volume = "13",
pages = "687--701",
journal = "IEEE Transactions on Medical Imaging",
issn = "0278-0062",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "4",

}

TY - JOUR

T1 - Fast gradient-based methods for Bayesian reconstruction of transmission and emission PET images

AU - Mumcuoglu, Erkan U.

AU - Leahy, Richard

AU - Cherry, Simon R.

AU - Cherry, Simon R

PY - 1994/12

Y1 - 1994/12

N2 - We describe conjugate gradient algorithms for reconstruction of transmission and emission PET images. The reconstructions are based on a Bayesian formulation, where the data are modeled as a collection of independent Poisson random variables and the image is modeled using a Markov random field. A conjugate gradient algorithm is used to compute a maximum a posteriori (MAP) estimate of the image by maximizing over the posterior density. To ensure nonnegativity of the solution, a penalty function is used to convert the problem to one of unconstrained optimization. Preconditioners are used to enhance convergence rates. These methods generally achieve effective convergence in 15-25 iterations. Reconstructions are presented of an 18FDG whole body scan from data collected using a Siemens/CTI ECAT931 whole body system. These results indicate significant improvements in emission image quality using the Bayesian approach, in comparison to filtered backprojection, particularly when reprojections of the MAP transmission image are used in place of the standard attenuation correction factors.

AB - We describe conjugate gradient algorithms for reconstruction of transmission and emission PET images. The reconstructions are based on a Bayesian formulation, where the data are modeled as a collection of independent Poisson random variables and the image is modeled using a Markov random field. A conjugate gradient algorithm is used to compute a maximum a posteriori (MAP) estimate of the image by maximizing over the posterior density. To ensure nonnegativity of the solution, a penalty function is used to convert the problem to one of unconstrained optimization. Preconditioners are used to enhance convergence rates. These methods generally achieve effective convergence in 15-25 iterations. Reconstructions are presented of an 18FDG whole body scan from data collected using a Siemens/CTI ECAT931 whole body system. These results indicate significant improvements in emission image quality using the Bayesian approach, in comparison to filtered backprojection, particularly when reprojections of the MAP transmission image are used in place of the standard attenuation correction factors.

UR - http://www.scopus.com/inward/record.url?scp=0028711560&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0028711560&partnerID=8YFLogxK

U2 - 10.1109/42.363099

DO - 10.1109/42.363099

M3 - Article

VL - 13

SP - 687

EP - 701

JO - IEEE Transactions on Medical Imaging

JF - IEEE Transactions on Medical Imaging

SN - 0278-0062

IS - 4

ER -