Maximum likelihood event estimation and list-mode image reconstruction on GPU hardware

Research output: Chapter in Book/Report/Conference proceedingConference contribution

14 Citations (Scopus)

Abstract

The scintillation detectors commonly used in SPECT and PET imaging and in Compton cameras require estimation of the position and energy of each gamma ray interaction. Ideally, this process would yield images with no spatial distortion and the best possible spatial resolution. In addition, especially for Compton cameras, the computation must yield the best possible estimate of the energy of each interacting gamma ray. These goals can be achieved by use of maximum-likelihood (ML) estimation of the event parameters, but in the past the search for an ML estimate has not been computationally feasible. Now, however, graphics processing units (GPUs) make it possible to produce optimal, real-time estimates of position and energy, even from scintillation cameras with a large number of photodetectors. In addition, the mathematical properties of ML estimates make them very attractive for use as list entries in list-mode ML image reconstruction. This two-step ML process - using ML estimation once to get the list data and again to reconstruct the object - allows accurate modeling of the detector blur and, potentially, considerable improvement in reconstructed spatial resolution.

Original languageEnglish (US)
Title of host publicationIEEE Nuclear Science Symposium Conference Record
Pages4072-4076
Number of pages5
DOIs
StatePublished - 2009
Event2009 IEEE Nuclear Science Symposium Conference Record, NSS/MIC 2009 - Orlando, FL, United States
Duration: Oct 25 2009Oct 31 2009

Other

Other2009 IEEE Nuclear Science Symposium Conference Record, NSS/MIC 2009
CountryUnited States
CityOrlando, FL
Period10/25/0910/31/09

Fingerprint

Likelihood Functions
Computer-Assisted Image Processing
Gamma Rays
image reconstruction
lists
hardware
maximum likelihood estimates
Gamma Cameras
cameras
Single-Photon Emission-Computed Tomography
scintillation
spatial resolution
gamma rays
detectors
estimates
entry
photometers
energy
interactions

ASJC Scopus subject areas

  • Radiation
  • Nuclear and High Energy Physics
  • Radiology Nuclear Medicine and imaging

Cite this

Maximum likelihood event estimation and list-mode image reconstruction on GPU hardware. / Caucci, Luca; Furenlid, Lars R; Barrett, Harrison H.

IEEE Nuclear Science Symposium Conference Record. 2009. p. 4072-4076 5402392.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Caucci, L, Furenlid, LR & Barrett, HH 2009, Maximum likelihood event estimation and list-mode image reconstruction on GPU hardware. in IEEE Nuclear Science Symposium Conference Record., 5402392, pp. 4072-4076, 2009 IEEE Nuclear Science Symposium Conference Record, NSS/MIC 2009, Orlando, FL, United States, 10/25/09. https://doi.org/10.1109/NSSMIC.2009.5402392
Caucci, Luca ; Furenlid, Lars R ; Barrett, Harrison H. / Maximum likelihood event estimation and list-mode image reconstruction on GPU hardware. IEEE Nuclear Science Symposium Conference Record. 2009. pp. 4072-4076
@inproceedings{d755a338621b4e958050c3c84db97d53,
title = "Maximum likelihood event estimation and list-mode image reconstruction on GPU hardware",
abstract = "The scintillation detectors commonly used in SPECT and PET imaging and in Compton cameras require estimation of the position and energy of each gamma ray interaction. Ideally, this process would yield images with no spatial distortion and the best possible spatial resolution. In addition, especially for Compton cameras, the computation must yield the best possible estimate of the energy of each interacting gamma ray. These goals can be achieved by use of maximum-likelihood (ML) estimation of the event parameters, but in the past the search for an ML estimate has not been computationally feasible. Now, however, graphics processing units (GPUs) make it possible to produce optimal, real-time estimates of position and energy, even from scintillation cameras with a large number of photodetectors. In addition, the mathematical properties of ML estimates make them very attractive for use as list entries in list-mode ML image reconstruction. This two-step ML process - using ML estimation once to get the list data and again to reconstruct the object - allows accurate modeling of the detector blur and, potentially, considerable improvement in reconstructed spatial resolution.",
author = "Luca Caucci and Furenlid, {Lars R} and Barrett, {Harrison H}",
year = "2009",
doi = "10.1109/NSSMIC.2009.5402392",
language = "English (US)",
isbn = "9781424439621",
pages = "4072--4076",
booktitle = "IEEE Nuclear Science Symposium Conference Record",

}

TY - GEN

T1 - Maximum likelihood event estimation and list-mode image reconstruction on GPU hardware

AU - Caucci, Luca

AU - Furenlid, Lars R

AU - Barrett, Harrison H

PY - 2009

Y1 - 2009

N2 - The scintillation detectors commonly used in SPECT and PET imaging and in Compton cameras require estimation of the position and energy of each gamma ray interaction. Ideally, this process would yield images with no spatial distortion and the best possible spatial resolution. In addition, especially for Compton cameras, the computation must yield the best possible estimate of the energy of each interacting gamma ray. These goals can be achieved by use of maximum-likelihood (ML) estimation of the event parameters, but in the past the search for an ML estimate has not been computationally feasible. Now, however, graphics processing units (GPUs) make it possible to produce optimal, real-time estimates of position and energy, even from scintillation cameras with a large number of photodetectors. In addition, the mathematical properties of ML estimates make them very attractive for use as list entries in list-mode ML image reconstruction. This two-step ML process - using ML estimation once to get the list data and again to reconstruct the object - allows accurate modeling of the detector blur and, potentially, considerable improvement in reconstructed spatial resolution.

AB - The scintillation detectors commonly used in SPECT and PET imaging and in Compton cameras require estimation of the position and energy of each gamma ray interaction. Ideally, this process would yield images with no spatial distortion and the best possible spatial resolution. In addition, especially for Compton cameras, the computation must yield the best possible estimate of the energy of each interacting gamma ray. These goals can be achieved by use of maximum-likelihood (ML) estimation of the event parameters, but in the past the search for an ML estimate has not been computationally feasible. Now, however, graphics processing units (GPUs) make it possible to produce optimal, real-time estimates of position and energy, even from scintillation cameras with a large number of photodetectors. In addition, the mathematical properties of ML estimates make them very attractive for use as list entries in list-mode ML image reconstruction. This two-step ML process - using ML estimation once to get the list data and again to reconstruct the object - allows accurate modeling of the detector blur and, potentially, considerable improvement in reconstructed spatial resolution.

UR - http://www.scopus.com/inward/record.url?scp=77951149222&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77951149222&partnerID=8YFLogxK

U2 - 10.1109/NSSMIC.2009.5402392

DO - 10.1109/NSSMIC.2009.5402392

M3 - Conference contribution

AN - SCOPUS:77951149222

SN - 9781424439621

SP - 4072

EP - 4076

BT - IEEE Nuclear Science Symposium Conference Record

ER -