Quantitative comparison of FBP, EM, and Bayesian reconstruction algorithms for the IndyPET scanner

Thomas Frese, Ned C. Rouze, Charles A. Bouman, Ken Sauer, Gary Hutchins

Research output: Contribution to journalArticle

64 Citations (Scopus)

Abstract

We quantitatively compare filtered backprojection (FBP), expectation-maximization (EM), and Bayesian reconstruction algorithms as applied to the IndyPET scanner - a dedicated research scanner which has been developed for small and intermediate field of view imaging applications. In contrast to previous approaches that rely on Monte Carlo simulations, a key feature of our investigation is the use of an empirical system kernel determined from scans of line source phantoms. This kernel is incorporated into the forward model of the EM and Bayesian algorithms to achieve resolution recovery. Three data sets are used, data collected on the IndyPET scanner using a bar phantom and a Hoffman three-dimensional brain phantom, and simulated data containing a hot lesion added to a uniform background. Reconstruction quality is analyzed quantitatively in terms of bias-variance measures (bar phantom) and mean square error (lesion phantom). We observe that without use of the empirical system kernel. the FBP, EM, and Bayesian algorithms give similar performance. However, with the inclusion of the empirical kernel, the iterative algorithms provide superior reconstructions compared with FBP, both in terms of visual quality and quantitative measures. Furthermore, Bayesian methods outperform EM. We conclude that significant improvements in reconstruction quality can be realized by combining accurate models of the system response with Bayesian reconstruction algorithms.

Original languageEnglish
Pages (from-to)258-276
Number of pages19
JournalIEEE Transactions on Medical Imaging
Volume22
Issue number2
DOIs
StatePublished - Feb 2003

Fingerprint

Bayes Theorem
Mean square error
Brain
Imaging techniques
Recovery
Research
Monte Carlo simulation
Datasets

Keywords

  • Bayesian methods
  • Image reconstruction
  • PET
  • Tomogrpahy

ASJC Scopus subject areas

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

Cite this

Quantitative comparison of FBP, EM, and Bayesian reconstruction algorithms for the IndyPET scanner. / Frese, Thomas; Rouze, Ned C.; Bouman, Charles A.; Sauer, Ken; Hutchins, Gary.

In: IEEE Transactions on Medical Imaging, Vol. 22, No. 2, 02.2003, p. 258-276.

Research output: Contribution to journalArticle

Frese, Thomas ; Rouze, Ned C. ; Bouman, Charles A. ; Sauer, Ken ; Hutchins, Gary. / Quantitative comparison of FBP, EM, and Bayesian reconstruction algorithms for the IndyPET scanner. In: IEEE Transactions on Medical Imaging. 2003 ; Vol. 22, No. 2. pp. 258-276.
@article{1cf0c0eb78f04692a7cc3c7e1c1817c7,
title = "Quantitative comparison of FBP, EM, and Bayesian reconstruction algorithms for the IndyPET scanner",
abstract = "We quantitatively compare filtered backprojection (FBP), expectation-maximization (EM), and Bayesian reconstruction algorithms as applied to the IndyPET scanner - a dedicated research scanner which has been developed for small and intermediate field of view imaging applications. In contrast to previous approaches that rely on Monte Carlo simulations, a key feature of our investigation is the use of an empirical system kernel determined from scans of line source phantoms. This kernel is incorporated into the forward model of the EM and Bayesian algorithms to achieve resolution recovery. Three data sets are used, data collected on the IndyPET scanner using a bar phantom and a Hoffman three-dimensional brain phantom, and simulated data containing a hot lesion added to a uniform background. Reconstruction quality is analyzed quantitatively in terms of bias-variance measures (bar phantom) and mean square error (lesion phantom). We observe that without use of the empirical system kernel. the FBP, EM, and Bayesian algorithms give similar performance. However, with the inclusion of the empirical kernel, the iterative algorithms provide superior reconstructions compared with FBP, both in terms of visual quality and quantitative measures. Furthermore, Bayesian methods outperform EM. We conclude that significant improvements in reconstruction quality can be realized by combining accurate models of the system response with Bayesian reconstruction algorithms.",
keywords = "Bayesian methods, Image reconstruction, PET, Tomogrpahy",
author = "Thomas Frese and Rouze, {Ned C.} and Bouman, {Charles A.} and Ken Sauer and Gary Hutchins",
year = "2003",
month = "2",
doi = "10.1109/TMI.2002.808353",
language = "English",
volume = "22",
pages = "258--276",
journal = "IEEE Transactions on Medical Imaging",
issn = "0278-0062",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "2",

}

TY - JOUR

T1 - Quantitative comparison of FBP, EM, and Bayesian reconstruction algorithms for the IndyPET scanner

AU - Frese, Thomas

AU - Rouze, Ned C.

AU - Bouman, Charles A.

AU - Sauer, Ken

AU - Hutchins, Gary

PY - 2003/2

Y1 - 2003/2

N2 - We quantitatively compare filtered backprojection (FBP), expectation-maximization (EM), and Bayesian reconstruction algorithms as applied to the IndyPET scanner - a dedicated research scanner which has been developed for small and intermediate field of view imaging applications. In contrast to previous approaches that rely on Monte Carlo simulations, a key feature of our investigation is the use of an empirical system kernel determined from scans of line source phantoms. This kernel is incorporated into the forward model of the EM and Bayesian algorithms to achieve resolution recovery. Three data sets are used, data collected on the IndyPET scanner using a bar phantom and a Hoffman three-dimensional brain phantom, and simulated data containing a hot lesion added to a uniform background. Reconstruction quality is analyzed quantitatively in terms of bias-variance measures (bar phantom) and mean square error (lesion phantom). We observe that without use of the empirical system kernel. the FBP, EM, and Bayesian algorithms give similar performance. However, with the inclusion of the empirical kernel, the iterative algorithms provide superior reconstructions compared with FBP, both in terms of visual quality and quantitative measures. Furthermore, Bayesian methods outperform EM. We conclude that significant improvements in reconstruction quality can be realized by combining accurate models of the system response with Bayesian reconstruction algorithms.

AB - We quantitatively compare filtered backprojection (FBP), expectation-maximization (EM), and Bayesian reconstruction algorithms as applied to the IndyPET scanner - a dedicated research scanner which has been developed for small and intermediate field of view imaging applications. In contrast to previous approaches that rely on Monte Carlo simulations, a key feature of our investigation is the use of an empirical system kernel determined from scans of line source phantoms. This kernel is incorporated into the forward model of the EM and Bayesian algorithms to achieve resolution recovery. Three data sets are used, data collected on the IndyPET scanner using a bar phantom and a Hoffman three-dimensional brain phantom, and simulated data containing a hot lesion added to a uniform background. Reconstruction quality is analyzed quantitatively in terms of bias-variance measures (bar phantom) and mean square error (lesion phantom). We observe that without use of the empirical system kernel. the FBP, EM, and Bayesian algorithms give similar performance. However, with the inclusion of the empirical kernel, the iterative algorithms provide superior reconstructions compared with FBP, both in terms of visual quality and quantitative measures. Furthermore, Bayesian methods outperform EM. We conclude that significant improvements in reconstruction quality can be realized by combining accurate models of the system response with Bayesian reconstruction algorithms.

KW - Bayesian methods

KW - Image reconstruction

KW - PET

KW - Tomogrpahy

UR - http://www.scopus.com/inward/record.url?scp=0038736677&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0038736677&partnerID=8YFLogxK

U2 - 10.1109/TMI.2002.808353

DO - 10.1109/TMI.2002.808353

M3 - Article

C2 - 12716002

AN - SCOPUS:0038736677

VL - 22

SP - 258

EP - 276

JO - IEEE Transactions on Medical Imaging

JF - IEEE Transactions on Medical Imaging

SN - 0278-0062

IS - 2

ER -