Jianan Cui1,2, Kuang Gong1,3, Ning Guo1,3, Chenxi Wu1, Xiaxia Meng1,4, Kyungsang Kim1,3, Kun Zheng5, Zhifang Wu4, Liping Fu6, Baixuan Xu6, Zhaohui Zhu5, Jiahe Tian6, Huafeng Liu7, Quanzheng Li8,9. 1. Center for Advanced Medical Computing and Analysis, Massachusetts General Hospital, 55 Fruit St, White 427, Boston, MA, 02114, USA. 2. State Key Laboratory of Modern Optical Instrumentation, College of Optical Science and Engineering, Zhejiang University, 38 Zheda Road, No.3 Teaching Building, 405, Hangzhou, 310027, China. 3. Gordon Center for Medical Imaging, Massachusetts General Hospital/Harvard Medical School, 55 Fruit St, White 427, Boston, MA, 02114, USA. 4. Department of Nuclear Medicine, First Hospital of Shanxi Medical University, Taiyuan, China. 5. Department of Nuclear Medicine, Peking Union Medical College Hospital, Beijing, China. 6. Department of Nuclear Medicine, The Chinese PLA General Hospital, Beijing, China. 7. State Key Laboratory of Modern Optical Instrumentation, College of Optical Science and Engineering, Zhejiang University, 38 Zheda Road, No.3 Teaching Building, 405, Hangzhou, 310027, China. liuhf@zju.edu.cn. 8. Center for Advanced Medical Computing and Analysis, Massachusetts General Hospital, 55 Fruit St, White 427, Boston, MA, 02114, USA. Li.Quanzheng@mgh.harvard.edu. 9. Gordon Center for Medical Imaging, Massachusetts General Hospital/Harvard Medical School, 55 Fruit St, White 427, Boston, MA, 02114, USA. Li.Quanzheng@mgh.harvard.edu.
Abstract
PURPOSE: Image quality of positron emission tomography (PET) is limited by various physical degradation factors. Our study aims to perform PET image denoising by utilizing prior information from the same patient. The proposed method is based on unsupervised deep learning, where no training pairs are needed. METHODS: In this method, the prior high-quality image from the patient was employed as the network input and the noisy PET image itself was treated as the training label. Constrained by the network structure and the prior image input, the network was trained to learn the intrinsic structure information from the noisy image and output a restored PET image. To validate the performance of the proposed method, a computer simulation study based on the BrainWeb phantom was first performed. A 68Ga-PRGD2 PET/CT dataset containing 10 patients and a 18F-FDG PET/MR dataset containing 30 patients were later on used for clinical data evaluation. The Gaussian, non-local mean (NLM) using CT/MR image as priors, BM4D, and Deep Decoder methods were included as reference methods. The contrast-to-noise ratio (CNR) improvements were used to rank different methods based on Wilcoxon signed-rank test. RESULTS: For the simulation study, contrast recovery coefficient (CRC) vs. standard deviation (STD) curves showed that the proposed method achieved the best performance regarding the bias-variance tradeoff. For the clinical PET/CT dataset, the proposed method achieved the highest CNR improvement ratio (53.35% ± 21.78%), compared with the Gaussian (12.64% ± 6.15%, P = 0.002), NLM guided by CT (24.35% ± 16.30%, P = 0.002), BM4D (38.31% ± 20.26%, P = 0.002), and Deep Decoder (41.67% ± 22.28%, P = 0.002) methods. For the clinical PET/MR dataset, the CNR improvement ratio of the proposed method achieved 46.80% ± 25.23%, higher than the Gaussian (18.16% ± 10.02%, P < 0.0001), NLM guided by MR (25.36% ± 19.48%, P < 0.0001), BM4D (37.02% ± 21.38%, P < 0.0001), and Deep Decoder (30.03% ± 20.64%, P < 0.0001) methods. Restored images for all the datasets demonstrate that the proposed method can effectively smooth out the noise while recovering image details. CONCLUSION: The proposed unsupervised deep learning framework provides excellent image restoration effects, outperforming the Gaussian, NLM methods, BM4D, and Deep Decoder methods.
PURPOSE: Image quality of positron emission tomography (PET) is limited by various physical degradation factors. Our study aims to perform PET image denoising by utilizing prior information from the same patient. The proposed method is based on unsupervised deep learning, where no training pairs are needed. METHODS: In this method, the prior high-quality image from the patient was employed as the network input and the noisy PET image itself was treated as the training label. Constrained by the network structure and the prior image input, the network was trained to learn the intrinsic structure information from the noisy image and output a restored PET image. To validate the performance of the proposed method, a computer simulation study based on the BrainWeb phantom was first performed. A 68Ga-PRGD2 PET/CT dataset containing 10 patients and a 18F-FDG PET/MR dataset containing 30 patients were later on used for clinical data evaluation. The Gaussian, non-local mean (NLM) using CT/MR image as priors, BM4D, and Deep Decoder methods were included as reference methods. The contrast-to-noise ratio (CNR) improvements were used to rank different methods based on Wilcoxon signed-rank test. RESULTS: For the simulation study, contrast recovery coefficient (CRC) vs. standard deviation (STD) curves showed that the proposed method achieved the best performance regarding the bias-variance tradeoff. For the clinical PET/CT dataset, the proposed method achieved the highest CNR improvement ratio (53.35% ± 21.78%), compared with the Gaussian (12.64% ± 6.15%, P = 0.002), NLM guided by CT (24.35% ± 16.30%, P = 0.002), BM4D (38.31% ± 20.26%, P = 0.002), and Deep Decoder (41.67% ± 22.28%, P = 0.002) methods. For the clinical PET/MR dataset, the CNR improvement ratio of the proposed method achieved 46.80% ± 25.23%, higher than the Gaussian (18.16% ± 10.02%, P < 0.0001), NLM guided by MR (25.36% ± 19.48%, P < 0.0001), BM4D (37.02% ± 21.38%, P < 0.0001), and Deep Decoder (30.03% ± 20.64%, P < 0.0001) methods. Restored images for all the datasets demonstrate that the proposed method can effectively smooth out the noise while recovering image details. CONCLUSION: The proposed unsupervised deep learning framework provides excellent image restoration effects, outperforming the Gaussian, NLM methods, BM4D, and Deep Decoder methods.
Entities:
Keywords:
Anatomical prior; Deep neural network; Denoising; Position emission tomography; Unsupervised deep learning
Authors: Claude Comtat; Paul E Kinahan; Jeffrey A Fessler; Thomas Beyer; David W Townsend; Michel Defrise; Christian Michel Journal: Phys Med Biol Date: 2002-01-07 Impact factor: 3.609
Authors: Tzu-An Song; Fan Yang; Samadrita Roy Chowdhury; Kyungsang Kim; Keith A Johnson; Georges El Fakhri; Quanzheng Li; Joyita Dutta Journal: IEEE Trans Comput Imaging Date: 2019-04-25
Authors: T Beyer; D W Townsend; T Brun; P E Kinahan; M Charron; R Roddy; J Jerin; J Young; L Byars; R Nutt Journal: J Nucl Med Date: 2000-08 Impact factor: 10.057