Statistical image reconstruction for low-dose CT using nonlocal means-based regularization

Hao Zhang, Jianhua Ma, Jing Wang, Yan Liu, Hongbing Lu, Zhengrong Liang

Research output: Contribution to journalArticle

39 Citations (Scopus)

Abstract

Low-dose computed tomography (CT) imaging without sacrifice of clinical tasks is desirable due to the growing concerns about excessive radiation exposure to the patients. One common strategy to achieve low-dose CT imaging is to lower the milliampere-second (mAs) setting in data scanning protocol. However, the reconstructed CT images by the conventional filtered back-projection (FBP) method from the low-mAs acquisitions may be severely degraded due to the excessive noise. Statistical image reconstruction (SIR) methods have shown potentials to significantly improve the reconstructed image quality from the low-mAs acquisitions, wherein the regularization plays a critical role and an established family of regularizations is based on the Markov random field (MRF) model. Inspired by the success of nonlocal means (NLM) in image processing applications, in this work, we propose to explore the NLM-based regularization for SIR to reconstruct low-dose CT images from low-mAs acquisitions. Experimental results with both digital and physical phantoms consistently demonstrated that SIR with the NLM-based regularization can achieve more gains than SIR with the well-known Gaussian MRF regularization or the generalized Gaussian MRF regularization and the conventional FBP method, in terms of image noise reduction and resolution preservation.

Original languageEnglish (US)
Pages (from-to)423-435
Number of pages13
JournalComputerized Medical Imaging and Graphics
Volume38
Issue number6
DOIs
StatePublished - 2014

Fingerprint

Computer-Assisted Image Processing
Image reconstruction
Tomography
Imaging techniques
Noise abatement
Image quality
Noise
Image processing
Scanning
Radiation

Keywords

  • Low-dose CT
  • Nonlocal means
  • Regularization
  • Statistical image reconstruction

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Computer Vision and Pattern Recognition
  • Radiology Nuclear Medicine and imaging
  • Health Informatics
  • Radiological and Ultrasound Technology

Cite this

Statistical image reconstruction for low-dose CT using nonlocal means-based regularization. / Zhang, Hao; Ma, Jianhua; Wang, Jing; Liu, Yan; Lu, Hongbing; Liang, Zhengrong.

In: Computerized Medical Imaging and Graphics, Vol. 38, No. 6, 2014, p. 423-435.

Research output: Contribution to journalArticle

Zhang, Hao ; Ma, Jianhua ; Wang, Jing ; Liu, Yan ; Lu, Hongbing ; Liang, Zhengrong. / Statistical image reconstruction for low-dose CT using nonlocal means-based regularization. In: Computerized Medical Imaging and Graphics. 2014 ; Vol. 38, No. 6. pp. 423-435.
@article{7c9b9043b0a146da84e698b47f92baed,
title = "Statistical image reconstruction for low-dose CT using nonlocal means-based regularization",
abstract = "Low-dose computed tomography (CT) imaging without sacrifice of clinical tasks is desirable due to the growing concerns about excessive radiation exposure to the patients. One common strategy to achieve low-dose CT imaging is to lower the milliampere-second (mAs) setting in data scanning protocol. However, the reconstructed CT images by the conventional filtered back-projection (FBP) method from the low-mAs acquisitions may be severely degraded due to the excessive noise. Statistical image reconstruction (SIR) methods have shown potentials to significantly improve the reconstructed image quality from the low-mAs acquisitions, wherein the regularization plays a critical role and an established family of regularizations is based on the Markov random field (MRF) model. Inspired by the success of nonlocal means (NLM) in image processing applications, in this work, we propose to explore the NLM-based regularization for SIR to reconstruct low-dose CT images from low-mAs acquisitions. Experimental results with both digital and physical phantoms consistently demonstrated that SIR with the NLM-based regularization can achieve more gains than SIR with the well-known Gaussian MRF regularization or the generalized Gaussian MRF regularization and the conventional FBP method, in terms of image noise reduction and resolution preservation.",
keywords = "Low-dose CT, Nonlocal means, Regularization, Statistical image reconstruction",
author = "Hao Zhang and Jianhua Ma and Jing Wang and Yan Liu and Hongbing Lu and Zhengrong Liang",
year = "2014",
doi = "10.1016/j.compmedimag.2014.05.002",
language = "English (US)",
volume = "38",
pages = "423--435",
journal = "Computerized Medical Imaging and Graphics",
issn = "0895-6111",
publisher = "Elsevier Limited",
number = "6",

}

TY - JOUR

T1 - Statistical image reconstruction for low-dose CT using nonlocal means-based regularization

AU - Zhang, Hao

AU - Ma, Jianhua

AU - Wang, Jing

AU - Liu, Yan

AU - Lu, Hongbing

AU - Liang, Zhengrong

PY - 2014

Y1 - 2014

N2 - Low-dose computed tomography (CT) imaging without sacrifice of clinical tasks is desirable due to the growing concerns about excessive radiation exposure to the patients. One common strategy to achieve low-dose CT imaging is to lower the milliampere-second (mAs) setting in data scanning protocol. However, the reconstructed CT images by the conventional filtered back-projection (FBP) method from the low-mAs acquisitions may be severely degraded due to the excessive noise. Statistical image reconstruction (SIR) methods have shown potentials to significantly improve the reconstructed image quality from the low-mAs acquisitions, wherein the regularization plays a critical role and an established family of regularizations is based on the Markov random field (MRF) model. Inspired by the success of nonlocal means (NLM) in image processing applications, in this work, we propose to explore the NLM-based regularization for SIR to reconstruct low-dose CT images from low-mAs acquisitions. Experimental results with both digital and physical phantoms consistently demonstrated that SIR with the NLM-based regularization can achieve more gains than SIR with the well-known Gaussian MRF regularization or the generalized Gaussian MRF regularization and the conventional FBP method, in terms of image noise reduction and resolution preservation.

AB - Low-dose computed tomography (CT) imaging without sacrifice of clinical tasks is desirable due to the growing concerns about excessive radiation exposure to the patients. One common strategy to achieve low-dose CT imaging is to lower the milliampere-second (mAs) setting in data scanning protocol. However, the reconstructed CT images by the conventional filtered back-projection (FBP) method from the low-mAs acquisitions may be severely degraded due to the excessive noise. Statistical image reconstruction (SIR) methods have shown potentials to significantly improve the reconstructed image quality from the low-mAs acquisitions, wherein the regularization plays a critical role and an established family of regularizations is based on the Markov random field (MRF) model. Inspired by the success of nonlocal means (NLM) in image processing applications, in this work, we propose to explore the NLM-based regularization for SIR to reconstruct low-dose CT images from low-mAs acquisitions. Experimental results with both digital and physical phantoms consistently demonstrated that SIR with the NLM-based regularization can achieve more gains than SIR with the well-known Gaussian MRF regularization or the generalized Gaussian MRF regularization and the conventional FBP method, in terms of image noise reduction and resolution preservation.

KW - Low-dose CT

KW - Nonlocal means

KW - Regularization

KW - Statistical image reconstruction

UR - http://www.scopus.com/inward/record.url?scp=84906064806&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84906064806&partnerID=8YFLogxK

U2 - 10.1016/j.compmedimag.2014.05.002

DO - 10.1016/j.compmedimag.2014.05.002

M3 - Article

C2 - 24881498

AN - SCOPUS:84906064806

VL - 38

SP - 423

EP - 435

JO - Computerized Medical Imaging and Graphics

JF - Computerized Medical Imaging and Graphics

SN - 0895-6111

IS - 6

ER -