Investigation on scale-based neighborhoods in MRFs for statistical iterative reconstruction

Hao Zhang, Yan Liu, Jing Wang, Jianhua Ma, Hao Han, Zhengrong Liang

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

Statistical iterative reconstruction (SIR) algorithms have shown advantages over the conventional filtered back-projection method for low-dose computed tomography (CT) reconstruction. For the SIR algorithms, the regularization term plays a critical role on determining the performance. One commonly used regularization is the quadratic-form Gaussian Markov random field (MRF), which penalizes differences among neighboring pixels in a small fixed window without considering discontinuities in images, thus may lead to over smoothing of edges or fine structures. In this work, we presented a quadratic-form MRF-based regularization with varying window size determined by the object scale, which is a descriptor of the image uniformity. For a uniform region (object scale is large), a larger MRF window is adopted because the coupling between the central pixel and its neighbors is strong; while for the interface region (object scale is small), a smaller MRF window is employed since the coupling is weak. The presented regularization term is incorporated into the penalized weighted least-squares (PWLS) iterative reconstruction scheme to improve low-dose CT reconstruction. Simulation results with a Shepp-Logan phantom revealed the presented regularization term is superior to the conventional Gaussian MRF in terms of noise suppression and edge preservation.

Original languageEnglish (US)
Title of host publicationIEEE Nuclear Science Symposium Conference Record
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Print)9781479905348
DOIs
StatePublished - 2013
Event2013 60th IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC 2013 - Seoul, Korea, Republic of
Duration: Oct 27 2013Nov 2 2013

Other

Other2013 60th IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC 2013
CountryKorea, Republic of
CitySeoul
Period10/27/1311/2/13

Fingerprint

Tomography
Least-Squares Analysis
Noise
tomography
pixels
dosage
smoothing
discontinuity
projection
fine structure
retarding
simulation

ASJC Scopus subject areas

  • Radiation
  • Nuclear and High Energy Physics
  • Radiology Nuclear Medicine and imaging

Cite this

Zhang, H., Liu, Y., Wang, J., Ma, J., Han, H., & Liang, Z. (2013). Investigation on scale-based neighborhoods in MRFs for statistical iterative reconstruction. In IEEE Nuclear Science Symposium Conference Record [6829374] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/NSSMIC.2013.6829374

Investigation on scale-based neighborhoods in MRFs for statistical iterative reconstruction. / Zhang, Hao; Liu, Yan; Wang, Jing; Ma, Jianhua; Han, Hao; Liang, Zhengrong.

IEEE Nuclear Science Symposium Conference Record. Institute of Electrical and Electronics Engineers Inc., 2013. 6829374.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Zhang, H, Liu, Y, Wang, J, Ma, J, Han, H & Liang, Z 2013, Investigation on scale-based neighborhoods in MRFs for statistical iterative reconstruction. in IEEE Nuclear Science Symposium Conference Record., 6829374, Institute of Electrical and Electronics Engineers Inc., 2013 60th IEEE Nuclear Science Symposium and Medical Imaging Conference, NSS/MIC 2013, Seoul, Korea, Republic of, 10/27/13. https://doi.org/10.1109/NSSMIC.2013.6829374
Zhang H, Liu Y, Wang J, Ma J, Han H, Liang Z. Investigation on scale-based neighborhoods in MRFs for statistical iterative reconstruction. In IEEE Nuclear Science Symposium Conference Record. Institute of Electrical and Electronics Engineers Inc. 2013. 6829374 https://doi.org/10.1109/NSSMIC.2013.6829374
Zhang, Hao ; Liu, Yan ; Wang, Jing ; Ma, Jianhua ; Han, Hao ; Liang, Zhengrong. / Investigation on scale-based neighborhoods in MRFs for statistical iterative reconstruction. IEEE Nuclear Science Symposium Conference Record. Institute of Electrical and Electronics Engineers Inc., 2013.
@inproceedings{0d931438ed104e399560fb582f8a81de,
title = "Investigation on scale-based neighborhoods in MRFs for statistical iterative reconstruction",
abstract = "Statistical iterative reconstruction (SIR) algorithms have shown advantages over the conventional filtered back-projection method for low-dose computed tomography (CT) reconstruction. For the SIR algorithms, the regularization term plays a critical role on determining the performance. One commonly used regularization is the quadratic-form Gaussian Markov random field (MRF), which penalizes differences among neighboring pixels in a small fixed window without considering discontinuities in images, thus may lead to over smoothing of edges or fine structures. In this work, we presented a quadratic-form MRF-based regularization with varying window size determined by the object scale, which is a descriptor of the image uniformity. For a uniform region (object scale is large), a larger MRF window is adopted because the coupling between the central pixel and its neighbors is strong; while for the interface region (object scale is small), a smaller MRF window is employed since the coupling is weak. The presented regularization term is incorporated into the penalized weighted least-squares (PWLS) iterative reconstruction scheme to improve low-dose CT reconstruction. Simulation results with a Shepp-Logan phantom revealed the presented regularization term is superior to the conventional Gaussian MRF in terms of noise suppression and edge preservation.",
author = "Hao Zhang and Yan Liu and Jing Wang and Jianhua Ma and Hao Han and Zhengrong Liang",
year = "2013",
doi = "10.1109/NSSMIC.2013.6829374",
language = "English (US)",
isbn = "9781479905348",
booktitle = "IEEE Nuclear Science Symposium Conference Record",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Investigation on scale-based neighborhoods in MRFs for statistical iterative reconstruction

AU - Zhang, Hao

AU - Liu, Yan

AU - Wang, Jing

AU - Ma, Jianhua

AU - Han, Hao

AU - Liang, Zhengrong

PY - 2013

Y1 - 2013

N2 - Statistical iterative reconstruction (SIR) algorithms have shown advantages over the conventional filtered back-projection method for low-dose computed tomography (CT) reconstruction. For the SIR algorithms, the regularization term plays a critical role on determining the performance. One commonly used regularization is the quadratic-form Gaussian Markov random field (MRF), which penalizes differences among neighboring pixels in a small fixed window without considering discontinuities in images, thus may lead to over smoothing of edges or fine structures. In this work, we presented a quadratic-form MRF-based regularization with varying window size determined by the object scale, which is a descriptor of the image uniformity. For a uniform region (object scale is large), a larger MRF window is adopted because the coupling between the central pixel and its neighbors is strong; while for the interface region (object scale is small), a smaller MRF window is employed since the coupling is weak. The presented regularization term is incorporated into the penalized weighted least-squares (PWLS) iterative reconstruction scheme to improve low-dose CT reconstruction. Simulation results with a Shepp-Logan phantom revealed the presented regularization term is superior to the conventional Gaussian MRF in terms of noise suppression and edge preservation.

AB - Statistical iterative reconstruction (SIR) algorithms have shown advantages over the conventional filtered back-projection method for low-dose computed tomography (CT) reconstruction. For the SIR algorithms, the regularization term plays a critical role on determining the performance. One commonly used regularization is the quadratic-form Gaussian Markov random field (MRF), which penalizes differences among neighboring pixels in a small fixed window without considering discontinuities in images, thus may lead to over smoothing of edges or fine structures. In this work, we presented a quadratic-form MRF-based regularization with varying window size determined by the object scale, which is a descriptor of the image uniformity. For a uniform region (object scale is large), a larger MRF window is adopted because the coupling between the central pixel and its neighbors is strong; while for the interface region (object scale is small), a smaller MRF window is employed since the coupling is weak. The presented regularization term is incorporated into the penalized weighted least-squares (PWLS) iterative reconstruction scheme to improve low-dose CT reconstruction. Simulation results with a Shepp-Logan phantom revealed the presented regularization term is superior to the conventional Gaussian MRF in terms of noise suppression and edge preservation.

UR - http://www.scopus.com/inward/record.url?scp=84897133672&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84897133672&partnerID=8YFLogxK

U2 - 10.1109/NSSMIC.2013.6829374

DO - 10.1109/NSSMIC.2013.6829374

M3 - Conference contribution

AN - SCOPUS:84897133672

SN - 9781479905348

BT - IEEE Nuclear Science Symposium Conference Record

PB - Institute of Electrical and Electronics Engineers Inc.

ER -