Image properties of various ML-based reconstructions of very noisy HRRT data

Simon Stute; Johan Nuyts; Katrien Van Slambrouck; Mérence Sibomana; Floris Van Velden; Ronald Boellaard; Claude Comtat

doi:10.1109/NSSMIC.2011.6153830

Image properties of various ML-based reconstructions of very noisy HRRT data

Simon Stute^*
, Johan Nuyts
, Katrien Van Slambrouck
, Mérence Sibomana
, Floris Van Velden
, Ronald Boellaard
, Claude Comtat

^*Corresponding author for this work

Service Hospitalier Frédéric Joliot
KU Leuven
Rigshospitalet

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

Abstract

The use of iterative image reconstruction algorithms with resolution modeling allows for reduced partial volume effect without noise increase. However, it is now recognized that EM-ML type algorithms are biased in very low counts images, in particular for cold regions. Alternative ML algorithms that allow for negative image voxels have been proposed to reduce the bias: NEG-ML of Nuyts et al and AB-ML of Byrne.

Original language	English
Title of host publication	2011 IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC 2011
Pages	4311-4315
Number of pages	5
DOIs	https://doi.org/10.1109/NSSMIC.2011.6153830
Publication status	Published - 26 Mar 2012
Event	2011 IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC 2011 - Duration: 26 Mar 2012 → …

Conference

Conference	2011 IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC 2011
Period	26/03/2012 → …

Access to Document

10.1109/NSSMIC.2011.6153830

Cite this

@inproceedings{9d267dabeef24c9c8c3c1e5472785ee2,

title = "Image properties of various ML-based reconstructions of very noisy HRRT data",

abstract = "The use of iterative image reconstruction algorithms with resolution modeling allows for reduced partial volume effect without noise increase. However, it is now recognized that EM-ML type algorithms are biased in very low counts images, in particular for cold regions. Alternative ML algorithms that allow for negative image voxels have been proposed to reduce the bias: NEG-ML of Nuyts et al and AB-ML of Byrne.",

author = "Simon Stute and Johan Nuyts and \{Van Slambrouck\}, Katrien and M{\'e}rence Sibomana and \{Van Velden\}, Floris and Ronald Boellaard and Claude Comtat",

year = "2012",

month = mar,

day = "26",

doi = "10.1109/NSSMIC.2011.6153830",

language = "English",

isbn = "9781467301183",

pages = "4311--4315",

booktitle = "2011 IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC 2011",

note = "2011 IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC 2011 ; Conference date: 26-03-2012",

}

Stute, S, Nuyts, J, Van Slambrouck, K, Sibomana, M, Van Velden, F, Boellaard, R & Comtat, C 2012, Image properties of various ML-based reconstructions of very noisy HRRT data. in 2011 IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC 2011., 6153830, pp. 4311-4315, 2011 IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC 2011, 26/03/2012. https://doi.org/10.1109/NSSMIC.2011.6153830

TY - GEN

T1 - Image properties of various ML-based reconstructions of very noisy HRRT data

AU - Stute, Simon

AU - Nuyts, Johan

AU - Van Slambrouck, Katrien

AU - Sibomana, Mérence

AU - Van Velden, Floris

AU - Boellaard, Ronald

AU - Comtat, Claude

PY - 2012/3/26

Y1 - 2012/3/26

N2 - The use of iterative image reconstruction algorithms with resolution modeling allows for reduced partial volume effect without noise increase. However, it is now recognized that EM-ML type algorithms are biased in very low counts images, in particular for cold regions. Alternative ML algorithms that allow for negative image voxels have been proposed to reduce the bias: NEG-ML of Nuyts et al and AB-ML of Byrne.

AB - The use of iterative image reconstruction algorithms with resolution modeling allows for reduced partial volume effect without noise increase. However, it is now recognized that EM-ML type algorithms are biased in very low counts images, in particular for cold regions. Alternative ML algorithms that allow for negative image voxels have been proposed to reduce the bias: NEG-ML of Nuyts et al and AB-ML of Byrne.

UR - https://www.scopus.com/pages/publications/84858658066

U2 - 10.1109/NSSMIC.2011.6153830

DO - 10.1109/NSSMIC.2011.6153830

M3 - Conference contribution

SN - 9781467301183

SP - 4311

EP - 4315

BT - 2011 IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC 2011

T2 - 2011 IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC 2011

Y2 - 26 March 2012

ER -

Image properties of various ML-based reconstructions of very noisy HRRT data

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Cite this