Optimal Nonlinear Line-of-Flight Estimation in Positron Emission Tomography

Alexander M. Bronstein; Michael M. Bronstein; Michael Zibulevsky; Yehoshua Y. Zeevi

doi:10.1109/TNS.2003.812434

Optimal Nonlinear Line-of-Flight Estimation in Positron Emission Tomography

Alexander M. Bronstein, Michael M. Bronstein, Michael Zibulevsky, Yehoshua Y. Zeevi

Research output: Contribution to journal › Article › peer-review

Abstract

The authors consider detection of high-energy photons in positron emission tomography using thick scintillation crystals. Parallax effect and multiple Compton interactions in such crystals significantly reduce the accuracy of conventional detection methods. In order to estimate the photon line of flight based on photomultiplier responses, the authors use asymptotically optimal nonlinear techniques, implemented by feed-forward and radial basis function neural networks. Incorporation of information about angles of incidence of photons significantly improves accuracy of estimation. The proposed estimators are fast enough to perform detection, using conventional computers. Monte Carlo simulation results show that their approach significantly outperforms the conventional Anger algorithm.

Original language	English
Pages (from-to)	421-426
Number of pages	6
Journal	IEEE Transactions on Nuclear Science
Volume	50
Issue number	3
DOIs	https://doi.org/10.1109/TNS.2003.812434
State	Published - Jun 2003
Externally published	Yes

Keywords

Artificial neural network
emission tomography
gamma camera
scintillation detector

Access to Document

10.1109/TNS.2003.812434

Cite this

@article{3d68a5bc74e0437fad0ac412fc4934b2,

title = "Optimal Nonlinear Line-of-Flight Estimation in Positron Emission Tomography",

abstract = "The authors consider detection of high-energy photons in positron emission tomography using thick scintillation crystals. Parallax effect and multiple Compton interactions in such crystals significantly reduce the accuracy of conventional detection methods. In order to estimate the photon line of flight based on photomultiplier responses, the authors use asymptotically optimal nonlinear techniques, implemented by feed-forward and radial basis function neural networks. Incorporation of information about angles of incidence of photons significantly improves accuracy of estimation. The proposed estimators are fast enough to perform detection, using conventional computers. Monte Carlo simulation results show that their approach significantly outperforms the conventional Anger algorithm.",

keywords = "Artificial neural network, emission tomography, gamma camera, scintillation detector",

author = "Bronstein, {Alexander M.} and Bronstein, {Michael M.} and Michael Zibulevsky and Zeevi, {Yehoshua Y.}",

note = "Funding Information: Manuscript received March 8, 2002. This work was supported in part by the Ollendorff Minerva Center, by the Fund for Promotion of Research at the Tech-nion, and by the Israeli Ministry of Science.",

year = "2003",

month = jun,

doi = "10.1109/TNS.2003.812434",

language = "אנגלית",

volume = "50",

pages = "421--426",

journal = "IEEE Transactions on Nuclear Science",

issn = "0018-9499",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "3",

}

TY - JOUR

T1 - Optimal Nonlinear Line-of-Flight Estimation in Positron Emission Tomography

AU - Bronstein, Alexander M.

AU - Bronstein, Michael M.

AU - Zibulevsky, Michael

AU - Zeevi, Yehoshua Y.

N1 - Funding Information: Manuscript received March 8, 2002. This work was supported in part by the Ollendorff Minerva Center, by the Fund for Promotion of Research at the Tech-nion, and by the Israeli Ministry of Science.

PY - 2003/6

Y1 - 2003/6

N2 - The authors consider detection of high-energy photons in positron emission tomography using thick scintillation crystals. Parallax effect and multiple Compton interactions in such crystals significantly reduce the accuracy of conventional detection methods. In order to estimate the photon line of flight based on photomultiplier responses, the authors use asymptotically optimal nonlinear techniques, implemented by feed-forward and radial basis function neural networks. Incorporation of information about angles of incidence of photons significantly improves accuracy of estimation. The proposed estimators are fast enough to perform detection, using conventional computers. Monte Carlo simulation results show that their approach significantly outperforms the conventional Anger algorithm.

AB - The authors consider detection of high-energy photons in positron emission tomography using thick scintillation crystals. Parallax effect and multiple Compton interactions in such crystals significantly reduce the accuracy of conventional detection methods. In order to estimate the photon line of flight based on photomultiplier responses, the authors use asymptotically optimal nonlinear techniques, implemented by feed-forward and radial basis function neural networks. Incorporation of information about angles of incidence of photons significantly improves accuracy of estimation. The proposed estimators are fast enough to perform detection, using conventional computers. Monte Carlo simulation results show that their approach significantly outperforms the conventional Anger algorithm.

KW - Artificial neural network

KW - emission tomography

KW - gamma camera

KW - scintillation detector

UR - http://www.scopus.com/inward/record.url?scp=0038382156&partnerID=8YFLogxK

U2 - 10.1109/TNS.2003.812434

DO - 10.1109/TNS.2003.812434

M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???

AN - SCOPUS:0038382156

VL - 50

SP - 421

EP - 426

JO - IEEE Transactions on Nuclear Science

JF - IEEE Transactions on Nuclear Science

SN - 0018-9499

IS - 3

ER -

Optimal Nonlinear Line-of-Flight Estimation in Positron Emission Tomography

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this