Iterative PET Image Reconstruction Using Convolutional Neural Network Representation

被引:194
作者
Gong, Kuang [1 ,2 ]
Guan, Jiahui [3 ]
Kim, Kyungsang [1 ]
Zhang, Xuezhu [2 ]
Yang, Jaewon [4 ]
Seo, Youngho [4 ]
El Fakhri, Georges [1 ]
Qi, Jinyi [2 ]
Li, Quanzheng [1 ]
机构
[1] Harvard Med Sch, Gordon Ctr Med Imaging, Massachusetts Gen Hosp, Boston, MA 02114 USA
[2] Univ Calif Davis, Dept Biomed Engn, Davis, CA 95616 USA
[3] Univ Calif Davis, Dept Stat, Davis, CA 95616 USA
[4] Univ Calif San Francisco, Phys Res Lab, Dept Radiol & Biomed Imaging, San Francisco, CA 94143 USA
基金
美国国家卫生研究院;
关键词
Positron emission tomography; convolutional neural network; iterative reconstruction; WHOLE-BODY PET; INFORMATION; TOMOGRAPHY; SCANNER; BRAIN;
D O I
10.1109/TMI.2018.2869871
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
PET image reconstruction is challenging due to the ill-poseness of the inverse problem and limited number of detected photons. Recently, the deep neural networks have been widely and successfully used in computer vision tasks and attracted growing interests in medical imaging. In this paper, we trained a deep residual convolutional neural network to improve PET image quality by using the existing inter-patient information. An innovative feature of the proposed method is that we embed the neural network in the iterative reconstruction framework for image representation, rather than using it as a post-processing tool. We formulate the objective function as a constrained optimization problem and solve it using the alternating direction method of multipliers algorithm. Both simulation data and hybrid real data are used to evaluate the proposed method. Quantification results show that our proposed iterative neural network method can outperform the neural network denoising and conventional penalized maximum likelihood methods.
引用
收藏
页码:675 / 685
页数:11
相关论文
共 55 条
[1]   Reconstruction of 3D Dynamic Contrast-Enhanced Magnetic Resonance Imaging Using Nonlocal Means [J].
Adluru, Ganesh ;
Tasdizen, Tolga ;
Schabel, Matthias C. ;
DiBella, Edward V. R. .
JOURNAL OF MAGNETIC RESONANCE IMAGING, 2010, 32 (05) :1217-1227
[2]   K-SVD: An algorithm for designing overcomplete dictionaries for sparse representation [J].
Aharon, Michal ;
Elad, Michael ;
Bruckstein, Alfred .
IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2006, 54 (11) :4311-4322
[3]  
[Anonymous], 2017, A cascaded convolutional neural network for X-ray low-dose CT image denoising
[4]  
[Anonymous], DEEP CONVOLUTIONAL N
[5]   Anatomical-based FDG-PET reconstruction for the detection of hypo-metabolic regions in epilepsy [J].
Baete, K ;
Nuyts, J ;
Van Paesschen, W ;
Suetens, P ;
Dupont, P .
IEEE TRANSACTIONS ON MEDICAL IMAGING, 2004, 23 (04) :510-519
[6]   Physical Performance of the new hybrid PET/CT Discovery-690 [J].
Bettinardi, V. ;
Presotto, L. ;
Rapisarda, E. ;
Picchio, M. ;
Gianolli, L. ;
Gilardi, M. C. .
MEDICAL PHYSICS, 2011, 38 (10) :5394-5411
[7]  
Beyer T, 2000, J NUCL MED, V41, P1369
[8]  
Bowsher JE, 2004, IEEE NUCL SCI CONF R, P2488
[9]   Distributed optimization and statistical learning via the alternating direction method of multipliers [J].
Boyd S. ;
Parikh N. ;
Chu E. ;
Peleato B. ;
Eckstein J. .
Foundations and Trends in Machine Learning, 2010, 3 (01) :1-122
[10]   Postreconstruction Nonlocal Means Filtering of Whole-Body PET With an Anatomical Prior [J].
Chan, Chung ;
Fulton, Roger ;
Barnett, Robert ;
Feng, David Dagan ;
Meikle, Steven .
IEEE TRANSACTIONS ON MEDICAL IMAGING, 2014, 33 (03) :636-650