Performance evaluation of MAP algorithms with different penalties, object geometries and noise levels

Yu Jung Tsai; Alexandre Bousse; Matthias J. Ehrhardt; Brian F. Hutton; Simon Arridge; Kris Thielemans

doi:10.1109/NSSMIC.2015.7582101

Performance evaluation of MAP algorithms with different penalties, object geometries and noise levels

Yu Jung Tsai, Alexandre Bousse, Matthias J. Ehrhardt, Brian F. Hutton, Simon Arridge, Kris Thielemans

Research output: Chapter or section in a book/report/conference proceeding › Chapter in a published conference proceeding

5 Citations (SciVal)

Abstract

A new algorithm (LBFGS-B-PC) which combines ideas of two existing convergent reconstruction algorithms, relaxed separable paraboloidal surrogate (SPS) and limited-memory Broyden-Fletcher-Goldfarb-Shanno with boundary constraints (LBFGS-B), is proposed. Its performance is evaluated in terms of log-posterior value and regional recovery ratio. The results demonstrate the superior convergence speed of the proposed algorithm to relaxed SPS and LBFGS-B, regardless of the noise level, activity distribution, object geometry, and penalties.

Original language	English
Title of host publication	2015 IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC
Place of Publication	U. S. A.
Publisher	IEEE
ISBN (Electronic)	9781467398626
DOIs	https://doi.org/10.1109/NSSMIC.2015.7582101
Publication status	Published - 3 Oct 2016
Event	2015 IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC 2015 - San Diego, USA United States Duration: 31 Oct 2015 → 7 Nov 2015

Conference

Conference	2015 IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC 2015
Country/Territory	USA United States
City	San Diego
Period	31/10/15 → 7/11/15

ASJC Scopus subject areas

Nuclear and High Energy Physics
Radiology Nuclear Medicine and imaging
Instrumentation

Access to Document

10.1109/NSSMIC.2015.7582101

Cite this

Performance evaluation of MAP algorithms with different penalties, object geometries and noise levels. / Tsai, Yu Jung; Bousse, Alexandre; Ehrhardt, Matthias J. et al.
2015 IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC. U. S. A.: IEEE, 2016. 7582101.

Research output: Chapter or section in a book/report/conference proceeding › Chapter in a published conference proceeding

Tsai, YJ, Bousse, A, Ehrhardt, MJ, Hutton, BF, Arridge, S & Thielemans, K 2016, Performance evaluation of MAP algorithms with different penalties, object geometries and noise levels. in 2015 IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC., 7582101, IEEE, U. S. A., 2015 IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC 2015, San Diego, USA United States, 31/10/15. https://doi.org/10.1109/NSSMIC.2015.7582101

@inproceedings{19383622b99e4daa8a108ab2b5eb1763,

title = "Performance evaluation of MAP algorithms with different penalties, object geometries and noise levels",

abstract = "A new algorithm (LBFGS-B-PC) which combines ideas of two existing convergent reconstruction algorithms, relaxed separable paraboloidal surrogate (SPS) and limited-memory Broyden-Fletcher-Goldfarb-Shanno with boundary constraints (LBFGS-B), is proposed. Its performance is evaluated in terms of log-posterior value and regional recovery ratio. The results demonstrate the superior convergence speed of the proposed algorithm to relaxed SPS and LBFGS-B, regardless of the noise level, activity distribution, object geometry, and penalties.",

author = "Tsai, {Yu Jung} and Alexandre Bousse and Ehrhardt, {Matthias J.} and Hutton, {Brian F.} and Simon Arridge and Kris Thielemans",

year = "2016",

month = oct,

day = "3",

doi = "10.1109/NSSMIC.2015.7582101",

language = "English",

booktitle = "2015 IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC",

publisher = "IEEE",

address = "USA United States",

note = "2015 IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC 2015 ; Conference date: 31-10-2015 Through 07-11-2015",

}

TY - GEN

T1 - Performance evaluation of MAP algorithms with different penalties, object geometries and noise levels

AU - Tsai, Yu Jung

AU - Bousse, Alexandre

AU - Ehrhardt, Matthias J.

AU - Hutton, Brian F.

AU - Arridge, Simon

AU - Thielemans, Kris

PY - 2016/10/3

Y1 - 2016/10/3

N2 - A new algorithm (LBFGS-B-PC) which combines ideas of two existing convergent reconstruction algorithms, relaxed separable paraboloidal surrogate (SPS) and limited-memory Broyden-Fletcher-Goldfarb-Shanno with boundary constraints (LBFGS-B), is proposed. Its performance is evaluated in terms of log-posterior value and regional recovery ratio. The results demonstrate the superior convergence speed of the proposed algorithm to relaxed SPS and LBFGS-B, regardless of the noise level, activity distribution, object geometry, and penalties.

AB - A new algorithm (LBFGS-B-PC) which combines ideas of two existing convergent reconstruction algorithms, relaxed separable paraboloidal surrogate (SPS) and limited-memory Broyden-Fletcher-Goldfarb-Shanno with boundary constraints (LBFGS-B), is proposed. Its performance is evaluated in terms of log-posterior value and regional recovery ratio. The results demonstrate the superior convergence speed of the proposed algorithm to relaxed SPS and LBFGS-B, regardless of the noise level, activity distribution, object geometry, and penalties.

UR - http://www.scopus.com/inward/record.url?scp=84994181923&partnerID=8YFLogxK

U2 - 10.1109/NSSMIC.2015.7582101

DO - 10.1109/NSSMIC.2015.7582101

M3 - Chapter in a published conference proceeding

AN - SCOPUS:84994181923

BT - 2015 IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC

PB - IEEE

CY - U. S. A.

T2 - 2015 IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC 2015

Y2 - 31 October 2015 through 7 November 2015

ER -

Performance evaluation of MAP algorithms with different penalties, object geometries and noise levels

Abstract

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this