Joint reconstruction of PET-MRI by parallel level sets

Matthias J. Ehrhardt; Kris Thielemans; Luis Pizarro; Pawel Markiewicz; David Atkinson; Sebastien Ourselin; Brian F. Hutton; Simon R. Arridge

doi:10.1109/NSSMIC.2014.7430895

Joint reconstruction of PET-MRI by parallel level sets

Matthias J. Ehrhardt, Kris Thielemans, Luis Pizarro, Pawel Markiewicz, David Atkinson, Sebastien Ourselin, Brian F. Hutton, Simon R. Arridge

Institute for Mathematical Innovation (IMI)

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

1 Citation (SciVal)

Abstract

Combined positron emission tomography (PET) and magnetic resonance imaging (MRI) scanners acquire simultaneously functional PET and anatomical or functional MRI data. As the data of both modalities are likely to show similar structures we aim to exploit this by joint reconstruction of PET and MRI. In a Bayesian formulation, this can be achieved by adding prior information encoding that the images of the two modalities are not independent. Structural similarity can be modeled by the alignment of the image gradients or equivalently their level sets being parallel. Therefore we can combine the objective functions of both modalities and penalize image pairs which do not have parallel level sets. Our results show that combining the reconstruction from heavily under-sampled MRI and noisy PET data can lead to less under-sampling artifacts in MRI images and better defined PET images.

Original language	English
Title of host publication	2014 IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC
Place of Publication	U. S. A.
Publisher	IEEE
ISBN (Electronic)	9781479960972
DOIs	https://doi.org/10.1109/NSSMIC.2014.7430895
Publication status	Published - 10 Mar 2016
Event	IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC 2014 - Seattle, USA United States Duration: 8 Nov 2014 → 15 Nov 2014

Conference

Conference	IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC 2014
Country/Territory	USA United States
City	Seattle
Period	8/11/14 → 15/11/14

Keywords

compressed sensing
image reconstruction
magnetic resonance imaging
positron emission tomography

ASJC Scopus subject areas

Nuclear and High Energy Physics
Radiology Nuclear Medicine and imaging

Access to Document

10.1109/NSSMIC.2014.7430895

Cite this

Ehrhardt, MJ, Thielemans, K, Pizarro, L, Markiewicz, P, Atkinson, D, Ourselin, S, Hutton, BF & Arridge, SR 2016, Joint reconstruction of PET-MRI by parallel level sets. in 2014 IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC ., 7430895, IEEE, U. S. A., IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC 2014, Seattle, USA United States, 8/11/14. https://doi.org/10.1109/NSSMIC.2014.7430895

@inproceedings{9da2a6ba8d984dcbb305ab2c13d15979,

title = "Joint reconstruction of PET-MRI by parallel level sets",

abstract = "Combined positron emission tomography (PET) and magnetic resonance imaging (MRI) scanners acquire simultaneously functional PET and anatomical or functional MRI data. As the data of both modalities are likely to show similar structures we aim to exploit this by joint reconstruction of PET and MRI. In a Bayesian formulation, this can be achieved by adding prior information encoding that the images of the two modalities are not independent. Structural similarity can be modeled by the alignment of the image gradients or equivalently their level sets being parallel. Therefore we can combine the objective functions of both modalities and penalize image pairs which do not have parallel level sets. Our results show that combining the reconstruction from heavily under-sampled MRI and noisy PET data can lead to less under-sampling artifacts in MRI images and better defined PET images.",

keywords = "compressed sensing, image reconstruction, magnetic resonance imaging, positron emission tomography",

author = "Ehrhardt, {Matthias J.} and Kris Thielemans and Luis Pizarro and Pawel Markiewicz and David Atkinson and Sebastien Ourselin and Hutton, {Brian F.} and Arridge, {Simon R.}",

year = "2016",

month = mar,

day = "10",

doi = "10.1109/NSSMIC.2014.7430895",

language = "English",

booktitle = "2014 IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC",

publisher = "IEEE",

address = "USA United States",

note = "IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC 2014 ; Conference date: 08-11-2014 Through 15-11-2014",

}

TY - GEN

T1 - Joint reconstruction of PET-MRI by parallel level sets

AU - Ehrhardt, Matthias J.

AU - Thielemans, Kris

AU - Pizarro, Luis

AU - Markiewicz, Pawel

AU - Atkinson, David

AU - Ourselin, Sebastien

AU - Hutton, Brian F.

AU - Arridge, Simon R.

PY - 2016/3/10

Y1 - 2016/3/10

N2 - Combined positron emission tomography (PET) and magnetic resonance imaging (MRI) scanners acquire simultaneously functional PET and anatomical or functional MRI data. As the data of both modalities are likely to show similar structures we aim to exploit this by joint reconstruction of PET and MRI. In a Bayesian formulation, this can be achieved by adding prior information encoding that the images of the two modalities are not independent. Structural similarity can be modeled by the alignment of the image gradients or equivalently their level sets being parallel. Therefore we can combine the objective functions of both modalities and penalize image pairs which do not have parallel level sets. Our results show that combining the reconstruction from heavily under-sampled MRI and noisy PET data can lead to less under-sampling artifacts in MRI images and better defined PET images.

AB - Combined positron emission tomography (PET) and magnetic resonance imaging (MRI) scanners acquire simultaneously functional PET and anatomical or functional MRI data. As the data of both modalities are likely to show similar structures we aim to exploit this by joint reconstruction of PET and MRI. In a Bayesian formulation, this can be achieved by adding prior information encoding that the images of the two modalities are not independent. Structural similarity can be modeled by the alignment of the image gradients or equivalently their level sets being parallel. Therefore we can combine the objective functions of both modalities and penalize image pairs which do not have parallel level sets. Our results show that combining the reconstruction from heavily under-sampled MRI and noisy PET data can lead to less under-sampling artifacts in MRI images and better defined PET images.

KW - compressed sensing

KW - image reconstruction

KW - magnetic resonance imaging

KW - positron emission tomography

UR - http://www.scopus.com/inward/record.url?scp=84965022893&partnerID=8YFLogxK

U2 - 10.1109/NSSMIC.2014.7430895

DO - 10.1109/NSSMIC.2014.7430895

M3 - Chapter in a published conference proceeding

AN - SCOPUS:84965022893

BT - 2014 IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC

PB - IEEE

CY - U. S. A.

T2 - IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC 2014

Y2 - 8 November 2014 through 15 November 2014

ER -

Joint reconstruction of PET-MRI by parallel level sets

Abstract

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this