Automated MAP-MRF em labelling for volume determination in PET

Hugh Gribben; Paul Miller; Hongbin Wang; Kathryn Carson; Alan Hounsell; Ashraf Zatari

doi:10.1109/ISBI.2008.4540917

Automated MAP-MRF em labelling for volume determination in PET

Hugh Gribben^*
, Paul Miller
, Hongbin Wang
, Kathryn Carson
, Alan Hounsell
, Ashraf Zatari

^*Corresponding author for this work

Queen's University Belfast

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

10 Citations (Scopus)

Abstract

An automated, unsupervised Maximum a Posterior - Markov Random Field Expectation Maximisation (MAP-MRF EM) Labelling technique, based upon a Bayesian framework, for volume of interest (VOI) determination in Positron Emission Tomography (PET) imagery is proposed. The segmentation technique incorporates MAP-MRF modelling into a mixture modelling approach using the EM algorithm, to consider both the structural and statistical nature of the data. The performance of the algorithm has been assessed on a set of PET phantom data. Investigations revealed improvements over a simple statistical approach using the EM algorithm, and improvements over a MAP-MRF approach, using the output from the EM algorithm as an initial estimate. Improvement is also shown over a standard semi-automated thresholding method, and an automated Fuzzy Hidden Markov Chain (FHMC) approach; particularly for smaller object volume determination, as the FHMC method loses some spatial correlation. A deblurring pre-processing stage was also found to provide improved results.

Original language	English
Title of host publication	2008 5th IEEE International Symposium on Biomedical Imaging
Subtitle of host publication	From Nano to Macro, Proceedings, ISBI
Pages	1-4
Number of pages	4
DOIs	https://doi.org/10.1109/ISBI.2008.4540917
Publication status	Published - 10 Sept 2008
Event	2008 5th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI - Paris, France Duration: 14 May 2008 → 17 May 2008

Conference

Conference	2008 5th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI
Country/Territory	France
City	Paris
Period	14/05/2008 → 17/05/2008

Keywords

EM
MAP-MRF
PET
Segmentation

ASJC Scopus subject areas

Biomedical Engineering

Access to Document

10.1109/ISBI.2008.4540917

Cite this

@inproceedings{d306023748404224b3a9586d1a503c9d,

title = "Automated MAP-MRF em labelling for volume determination in PET",

abstract = "An automated, unsupervised Maximum a Posterior - Markov Random Field Expectation Maximisation (MAP-MRF EM) Labelling technique, based upon a Bayesian framework, for volume of interest (VOI) determination in Positron Emission Tomography (PET) imagery is proposed. The segmentation technique incorporates MAP-MRF modelling into a mixture modelling approach using the EM algorithm, to consider both the structural and statistical nature of the data. The performance of the algorithm has been assessed on a set of PET phantom data. Investigations revealed improvements over a simple statistical approach using the EM algorithm, and improvements over a MAP-MRF approach, using the output from the EM algorithm as an initial estimate. Improvement is also shown over a standard semi-automated thresholding method, and an automated Fuzzy Hidden Markov Chain (FHMC) approach; particularly for smaller object volume determination, as the FHMC method loses some spatial correlation. A deblurring pre-processing stage was also found to provide improved results.",

keywords = "EM, MAP-MRF, PET, Segmentation",

author = "Hugh Gribben and Paul Miller and Hongbin Wang and Kathryn Carson and Alan Hounsell and Ashraf Zatari",

year = "2008",

month = sep,

day = "10",

doi = "10.1109/ISBI.2008.4540917",

language = "English",

isbn = "9781424420032",

pages = "1--4",

booktitle = "2008 5th IEEE International Symposium on Biomedical Imaging",

note = "2008 5th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI ; Conference date: 14-05-2008 Through 17-05-2008",

}

Gribben, H, Miller, P, Wang, H, Carson, K, Hounsell, A & Zatari, A 2008, Automated MAP-MRF em labelling for volume determination in PET. in 2008 5th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, Proceedings, ISBI., 4540917, pp. 1-4, 2008 5th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI, Paris, France, 14/05/2008. https://doi.org/10.1109/ISBI.2008.4540917

TY - GEN

T1 - Automated MAP-MRF em labelling for volume determination in PET

AU - Gribben, Hugh

AU - Miller, Paul

AU - Wang, Hongbin

AU - Carson, Kathryn

AU - Hounsell, Alan

AU - Zatari, Ashraf

PY - 2008/9/10

Y1 - 2008/9/10

N2 - An automated, unsupervised Maximum a Posterior - Markov Random Field Expectation Maximisation (MAP-MRF EM) Labelling technique, based upon a Bayesian framework, for volume of interest (VOI) determination in Positron Emission Tomography (PET) imagery is proposed. The segmentation technique incorporates MAP-MRF modelling into a mixture modelling approach using the EM algorithm, to consider both the structural and statistical nature of the data. The performance of the algorithm has been assessed on a set of PET phantom data. Investigations revealed improvements over a simple statistical approach using the EM algorithm, and improvements over a MAP-MRF approach, using the output from the EM algorithm as an initial estimate. Improvement is also shown over a standard semi-automated thresholding method, and an automated Fuzzy Hidden Markov Chain (FHMC) approach; particularly for smaller object volume determination, as the FHMC method loses some spatial correlation. A deblurring pre-processing stage was also found to provide improved results.

AB - An automated, unsupervised Maximum a Posterior - Markov Random Field Expectation Maximisation (MAP-MRF EM) Labelling technique, based upon a Bayesian framework, for volume of interest (VOI) determination in Positron Emission Tomography (PET) imagery is proposed. The segmentation technique incorporates MAP-MRF modelling into a mixture modelling approach using the EM algorithm, to consider both the structural and statistical nature of the data. The performance of the algorithm has been assessed on a set of PET phantom data. Investigations revealed improvements over a simple statistical approach using the EM algorithm, and improvements over a MAP-MRF approach, using the output from the EM algorithm as an initial estimate. Improvement is also shown over a standard semi-automated thresholding method, and an automated Fuzzy Hidden Markov Chain (FHMC) approach; particularly for smaller object volume determination, as the FHMC method loses some spatial correlation. A deblurring pre-processing stage was also found to provide improved results.

KW - EM

KW - MAP-MRF

KW - PET

KW - Segmentation

UR - https://www.scopus.com/pages/publications/51049109213

U2 - 10.1109/ISBI.2008.4540917

DO - 10.1109/ISBI.2008.4540917

M3 - Conference contribution

AN - SCOPUS:51049109213

SN - 9781424420032

SP - 1

EP - 4

BT - 2008 5th IEEE International Symposium on Biomedical Imaging

T2 - 2008 5th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI

Y2 - 14 May 2008 through 17 May 2008

ER -

Automated MAP-MRF em labelling for volume determination in PET

Abstract

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this