EM-ML PET reconstruction on multiple processors with reduced communications

Søren P. Olesen, Jens Gregor, Michael G. Thomason, Gary T. Smith

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

Positron emission tomography (PET) reconstruction by the EM algorithm is an iterative computation of Poisson emission rates to maximize a likelihood function. The method is time consuming and, for real scanner data, requires large numerical arrays. To speed up the computation on multiple processors which have their own local memory and communicate by passing messages on a network, a parallel method has been implemented in which processors compute several iterations before exchanging their latest data with other processors. This method is convenient for iterative reconstruction using a relatively small number of interconnected, standard processors such as workstations on a local-area network. Computational aspects of the method are explained and illustrated with two-dimensional reconstructions from a simulation and from sinograms produced by a PET scanner. Five hundred twelve (512) iterations are computed on a local-area network of workstations and, for reference, on a distributed-memory multiprocessor computer. The method is capable of producing high-quality reconstructions with significant speed-up.

Original languageEnglish (US)
Pages (from-to)215-223
Number of pages9
JournalInternational Journal of Imaging Systems and Technology
Volume7
Issue number3
DOIs
StatePublished - Jan 1 1996
Externally publishedYes

Fingerprint

Positron emission tomography
Computer workstations
Local area networks
Data storage equipment
Communication
Message passing

All Science Journal Classification (ASJC) codes

  • Electronic, Optical and Magnetic Materials
  • Software
  • Computer Vision and Pattern Recognition
  • Electrical and Electronic Engineering

Cite this

EM-ML PET reconstruction on multiple processors with reduced communications. / Olesen, Søren P.; Gregor, Jens; Thomason, Michael G.; Smith, Gary T.

In: International Journal of Imaging Systems and Technology, Vol. 7, No. 3, 01.01.1996, p. 215-223.

Research output: Contribution to journalArticle

Olesen, Søren P. ; Gregor, Jens ; Thomason, Michael G. ; Smith, Gary T. / EM-ML PET reconstruction on multiple processors with reduced communications. In: International Journal of Imaging Systems and Technology. 1996 ; Vol. 7, No. 3. pp. 215-223.
@article{63011f57bd1a40519157f5c7e69ba14f,
title = "EM-ML PET reconstruction on multiple processors with reduced communications",
abstract = "Positron emission tomography (PET) reconstruction by the EM algorithm is an iterative computation of Poisson emission rates to maximize a likelihood function. The method is time consuming and, for real scanner data, requires large numerical arrays. To speed up the computation on multiple processors which have their own local memory and communicate by passing messages on a network, a parallel method has been implemented in which processors compute several iterations before exchanging their latest data with other processors. This method is convenient for iterative reconstruction using a relatively small number of interconnected, standard processors such as workstations on a local-area network. Computational aspects of the method are explained and illustrated with two-dimensional reconstructions from a simulation and from sinograms produced by a PET scanner. Five hundred twelve (512) iterations are computed on a local-area network of workstations and, for reference, on a distributed-memory multiprocessor computer. The method is capable of producing high-quality reconstructions with significant speed-up.",
author = "Olesen, {S{\o}ren P.} and Jens Gregor and Thomason, {Michael G.} and Smith, {Gary T.}",
year = "1996",
month = "1",
day = "1",
doi = "10.1002/(SICI)1098-1098(199623)7:3<215::AID-IMA8>3.0.CO;2-B",
language = "English (US)",
volume = "7",
pages = "215--223",
journal = "International Journal of Imaging Systems and Technology",
issn = "0899-9457",
publisher = "John Wiley and Sons Inc.",
number = "3",

}

TY - JOUR

T1 - EM-ML PET reconstruction on multiple processors with reduced communications

AU - Olesen, Søren P.

AU - Gregor, Jens

AU - Thomason, Michael G.

AU - Smith, Gary T.

PY - 1996/1/1

Y1 - 1996/1/1

N2 - Positron emission tomography (PET) reconstruction by the EM algorithm is an iterative computation of Poisson emission rates to maximize a likelihood function. The method is time consuming and, for real scanner data, requires large numerical arrays. To speed up the computation on multiple processors which have their own local memory and communicate by passing messages on a network, a parallel method has been implemented in which processors compute several iterations before exchanging their latest data with other processors. This method is convenient for iterative reconstruction using a relatively small number of interconnected, standard processors such as workstations on a local-area network. Computational aspects of the method are explained and illustrated with two-dimensional reconstructions from a simulation and from sinograms produced by a PET scanner. Five hundred twelve (512) iterations are computed on a local-area network of workstations and, for reference, on a distributed-memory multiprocessor computer. The method is capable of producing high-quality reconstructions with significant speed-up.

AB - Positron emission tomography (PET) reconstruction by the EM algorithm is an iterative computation of Poisson emission rates to maximize a likelihood function. The method is time consuming and, for real scanner data, requires large numerical arrays. To speed up the computation on multiple processors which have their own local memory and communicate by passing messages on a network, a parallel method has been implemented in which processors compute several iterations before exchanging their latest data with other processors. This method is convenient for iterative reconstruction using a relatively small number of interconnected, standard processors such as workstations on a local-area network. Computational aspects of the method are explained and illustrated with two-dimensional reconstructions from a simulation and from sinograms produced by a PET scanner. Five hundred twelve (512) iterations are computed on a local-area network of workstations and, for reference, on a distributed-memory multiprocessor computer. The method is capable of producing high-quality reconstructions with significant speed-up.

UR - http://www.scopus.com/inward/record.url?scp=0030231251&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030231251&partnerID=8YFLogxK

U2 - 10.1002/(SICI)1098-1098(199623)7:3<215::AID-IMA8>3.0.CO;2-B

DO - 10.1002/(SICI)1098-1098(199623)7:3<215::AID-IMA8>3.0.CO;2-B

M3 - Article

VL - 7

SP - 215

EP - 223

JO - International Journal of Imaging Systems and Technology

JF - International Journal of Imaging Systems and Technology

SN - 0899-9457

IS - 3

ER -