Parameter-Transferred Wasserstein Generative Adversarial Network (PT-WGAN) for Low-Dose PET Image Denoising

Due to the widespread use of positron emission tomography (PET) in clinical practice, the potential risk of PET-associated radiation dose to patients needs to be minimized. However, with the reduction in the radiation dose, the resultant images may suffer from noise and artifacts that compromise diagnostic performance. In this paper, we propose a parameter-transferred Wasserstein generative adversarial network (PT-WGAN) for low-dose PET image denoising. The contributions of this paper are twofold: i) a PT-WGAN framework is designed to denoise low-dose PET images without compromising structural details, and ii) a task-specific initialization based on transfer learning is developed to train PT-WGAN using trainable parameters transferred from a pretrained model, which significantly improves the training efficiency of PT-WGAN. The experimental results on clinical data show that the proposed network can suppress image noise more effectively while preserving better image fidelity than recently published state-of-the-art methods. We make our code available at this https URL.

[1]  Phaneendra K. Yalavarthy,et al.  Convolutional Neural Network-Based Robust Denoising of Low-Dose Computed Tomography Perfusion Maps , 2019, IEEE Transactions on Radiation and Plasma Medical Sciences.

[2]  Jean-Michel Morel,et al.  A non-local algorithm for image denoising , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[3]  Uwe Kruger,et al.  Competitive performance of a modularized deep neural network compared to commercial algorithms for low-dose CT image reconstruction , 2019, Nat. Mach. Intell..

[4]  Yu-Bin Yang,et al.  Image Restoration Using Very Deep Convolutional Encoder-Decoder Networks with Symmetric Skip Connections , 2016, NIPS.

[5]  Qiegen Liu,et al.  Model-based Convolutional De-Aliasing Network Learning for Parallel MR Imaging , 2019, MICCAI.

[6]  Dinggang Shen,et al.  3D conditional generative adversarial networks for high-quality PET image estimation at low dose , 2018, NeuroImage.

[7]  F. Noo,et al.  A sequential solution for anisotropic total variation image denoising with interval constraints , 2017, Physics in medicine and biology.

[8]  Yun Liang,et al.  STIR-Net: Deep Spatial-Temporal Image Restoration Net for Radiation Reduction in CT Perfusion , 2019, Front. Neurol..

[9]  Junshen Xu,et al.  200x Low-dose PET Reconstruction using Deep Learning , 2017, ArXiv.

[10]  Huafeng Liu,et al.  Low dose PET reconstruction with total variation regularization , 2014, 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[11]  Hairong Zheng,et al.  CLCI-Net: Cross-Level fusion and Context Inference Networks for Lesion Segmentation of Chronic Stroke , 2019, MICCAI.

[12]  John M Pauly,et al.  Ultra-Low-Dose 18F-Florbetaben Amyloid PET Imaging Using Deep Learning with Multi-Contrast MRI Inputs. , 2019, Radiology.

[13]  K. Erlandsson,et al.  Fast accurate iterative reconstruction for low-statistics positron volume imaging. , 1998, Physics in medicine and biology.

[14]  Dong Liang,et al.  DeepcomplexMRI: Exploiting deep residual network for fast parallel MR imaging with complex convolution. , 2020, Magnetic resonance imaging.

[15]  Tianwei Ni,et al.  Elastic Boundary Projection for 3D Medical Image Segmentation , 2018, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[16]  Paul Kinahan,et al.  A combined PET/CT scanner for clinical oncology. , 2000, Journal of nuclear medicine : official publication, Society of Nuclear Medicine.

[17]  Martín Abadi,et al.  TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems , 2016, ArXiv.

[18]  Seyed-Ahmad Ahmadi,et al.  V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation , 2016, 2016 Fourth International Conference on 3D Vision (3DV).

[19]  Feng Lin,et al.  Low-Dose CT With a Residual Encoder-Decoder Convolutional Neural Network , 2017, IEEE Transactions on Medical Imaging.

[20]  Quanzheng Li,et al.  MAPEM-Net: an unrolled neural network for Fully 3D PET image reconstruction , 2019 .

[21]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[22]  Sun Il Kwon,et al.  Signal encoding method for a time-of-flight PET detector using a silicon photomultiplier array , 2014 .

[23]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[24]  Xiang Li,et al.  Deep Learning-Based Image Segmentation on Multimodal Medical Imaging , 2019, IEEE Transactions on Radiation and Plasma Medical Sciences.

[25]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[26]  Yoshua Bengio,et al.  Generative Adversarial Nets , 2014, NIPS.

[27]  Paul Babyn,et al.  Sharpness-Aware Low-Dose CT Denoising Using Conditional Generative Adversarial Network , 2017, Journal of Digital Imaging.

[28]  Hongming Shan,et al.  3-D Convolutional Encoder-Decoder Network for Low-Dose CT via Transfer Learning From a 2-D Trained Network , 2018, IEEE Transactions on Medical Imaging.

[29]  Fan Yang,et al.  PET Image Deblurring and Super-Resolution With an MR-Based Joint Entropy Prior , 2019, IEEE Transactions on Computational Imaging.

[30]  Guobao Wang,et al.  Penalized Likelihood PET Image Reconstruction Using Patch-Based Edge-Preserving Regularization , 2012, IEEE Transactions on Medical Imaging.

[31]  Yoshua Bengio,et al.  Understanding the difficulty of training deep feedforward neural networks , 2010, AISTATS.

[32]  F. Hashimoto,et al.  Denoising of Dynamic Sinogram by Image Guided Filtering for Positron Emission Tomography , 2018, IEEE Transactions on Radiation and Plasma Medical Sciences.

[33]  Mark Slifstein,et al.  Quantitative imaging of protein targets in the human brain with PET , 2015, Physics in medicine and biology.

[34]  Quanzheng Li,et al.  Attenuation Correction of PET/MR Using Cycle-Consistent Adversarial Network , 2019 .

[35]  Lijun Zhao,et al.  Automatic Nasopharyngeal Carcinoma Segmentation Using Fully Convolutional Networks with Auxiliary Paths on Dual-Modality PET-CT Images , 2019, Journal of Digital Imaging.

[36]  Alan C. Bovik,et al.  Image information and visual quality , 2006, IEEE Trans. Image Process..

[37]  Josef Machac,et al.  Cardiac positron emission tomography imaging. , 2005, Seminars in nuclear medicine.

[38]  Léon Bottou,et al.  Wasserstein Generative Adversarial Networks , 2017, ICML.

[39]  Aaron C. Courville,et al.  Improved Training of Wasserstein GANs , 2017, NIPS.

[40]  Quanzheng Li,et al.  Iterative PET Image Reconstruction Using Convolutional Neural Network Representation , 2017, IEEE Transactions on Medical Imaging.

[41]  Kuang Gong,et al.  Generative adversarial network based regularized image reconstruction for PET , 2019, Physics in medicine and biology.

[42]  Lei Zhang,et al.  RFSIM: A feature based image quality assessment metric using Riesz transforms , 2010, 2010 IEEE International Conference on Image Processing.

[43]  Leslie Ying,et al.  Accelerating magnetic resonance imaging via deep learning , 2016, 2016 IEEE 13th International Symposium on Biomedical Imaging (ISBI).

[44]  Khan A. Wahid,et al.  Denoising Low-Dose CT Images Using Multiframe Blind Source Separation and Block Matching Filter , 2018, IEEE Transactions on Radiation and Plasma Medical Sciences.

[45]  Dinggang Shen,et al.  Deep auto-context convolutional neural networks for standard-dose PET image estimation from low-dose PET/MRI , 2017, Neurocomputing.

[46]  Christian Barillot,et al.  The first MICCAI challenge on PET tumor segmentation , 2018, Medical Image Anal..

[47]  Eero P. Simoncelli,et al.  Image quality assessment: from error visibility to structural similarity , 2004, IEEE Transactions on Image Processing.

[48]  Qiang Yang,et al.  A Survey on Transfer Learning , 2010, IEEE Transactions on Knowledge and Data Engineering.

[49]  Chih-Chieh Liu,et al.  PET Image Denoising Using a Deep Neural Network Through Fine Tuning , 2019, IEEE Transactions on Radiation and Plasma Medical Sciences.

[50]  Pawel Markiewicz,et al.  Faster PET reconstruction with non-smooth priors by randomization and preconditioning , 2018, Physics in medicine and biology.