A Model-Based Unsupervised Deep Learning Method for Low-Dose CT Reconstruction

Low-dose CT (LDCT) is of great significance due to the concern about the potential radiation risk. With the fast development of deep learning, neural networks have become powerful tools in LDCT enhancement. Current deep neural networks for LDCT reconstruction are often trained with paired LDCT dataset and normal-dose CT (NDCT) dataset. However, high quality NDCT dataset paired with LDCT dataset is expensive to acquire or even not available sometimes in reality. In this work, we proposed an unsupervised model-based deep learning (MBDL) for LDCT reconstruction. The network is trained based on group-wise maximum a posterior (G-MAP) loss function with LDCT dataset only. The MBDL is a general framework. It also allows us to combine with supervised training if a small number of paired NDCT dataset accessible to help optimizing the network parameters, i.e. works in a semi-supervised mode. During inference, LDCT images are reconstructed end-to-end by the trained network. We verified the proposed method with simulated projection data from clinical CT images. The proposed method restrained noise well while restoring anatomical structures and it achieved better results than model-based iterative reconstruction (MBIR) with significantly less computational cost. The performances of MBDL were further enhanced by integrating a small paired NDCT dataset for semi-supervised training. The results suggested that MBDL is an efficient and flexible method for LDCT deep learning based reconstruction in the situations lacking of enough high quality NDCT data.

[1]  Ken D. Sauer,et al.  A Gaussian Mixture MRF for Model-Based Iterative Reconstruction With Applications to Low-Dose X-Ray CT , 2016, IEEE Transactions on Computational Imaging.

[2]  Loïc Royer,et al.  Noise2Self: Blind Denoising by Self-Supervision , 2019, ICML.

[3]  Javad Alirezaie,et al.  Low-dose CT Denoising with Dilated Residual Network , 2018, 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).

[4]  Junyan Rong,et al.  Low-Dose Lung CT Image Restoration Using Adaptive Prior Features From Full-Dose Training Database , 2017, IEEE Transactions on Medical Imaging.

[5]  Lei Zhang,et al.  Low-Dose X-ray CT Reconstruction via Dictionary Learning , 2012, IEEE Transactions on Medical Imaging.

[6]  Zhengrong Liang,et al.  An experimental study on the noise properties of x-ray CT sinogram data in Radon space , 2008, Physics in medicine and biology.

[7]  Yan Liu,et al.  Low-dose CT restoration via stacked sparse denoising autoencoders , 2018, Neurocomputing.

[8]  Jong Chul Ye,et al.  Framing U-Net via Deep Convolutional Framelets: Application to Sparse-View CT , 2017, IEEE Transactions on Medical Imaging.

[9]  Jingyan Xu,et al.  Electronic Noise Modeling in Statistical Iterative Reconstruction , 2009, IEEE Transactions on Image Processing.

[10]  Jonas Adler,et al.  Learned Primal-Dual Reconstruction , 2017, IEEE Transactions on Medical Imaging.

[11]  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.

[12]  Uwe Kruger,et al.  A novel transfer learning framework for low-dose CT , 2019, 15th International Meeting on Fully Three-Dimensional Image Reconstruction in Radiology and Nuclear Medicine.

[13]  Yurii Nesterov,et al.  Efficiency of Coordinate Descent Methods on Huge-Scale Optimization Problems , 2012, SIAM J. Optim..

[14]  D. Brenner,et al.  Computed tomography--an increasing source of radiation exposure. , 2007, The New England journal of medicine.

[15]  Jaakko Lehtinen,et al.  Noise2Noise: Learning Image Restoration without Clean Data , 2018, ICML.

[16]  Ke Sheng,et al.  Denoising of low dose CT image with context-based BM3D , 2016, 2016 IEEE Region 10 Conference (TENCON).

[17]  Jing Wang,et al.  Statistical image reconstruction for low-dose CT using nonlocal means-based regularization , 2014, Comput. Medical Imaging Graph..

[18]  Max A. Viergever,et al.  Generative Adversarial Networks for Noise Reduction in Low-Dose CT , 2017, IEEE Transactions on Medical Imaging.

[19]  Jong Chul Ye,et al.  Low-dose limited view 4D CT reconstruction using patch-based low-rank regularization , 2013, 2013 IEEE Nuclear Science Symposium and Medical Imaging Conference (2013 NSS/MIC).

[20]  Florian Jug,et al.  Noise2Void - Learning Denoising From Single Noisy Images , 2018, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[21]  R. Siddon.  Fast calculation of the exact radiological path for a three-dimensional CT array. , 1985, Medical physics.

[22]  Jin Liu,et al.  Low-dose CBCT reconstruction via 3D dictionary learning , 2016, 2016 IEEE 13th International Symposium on Biomedical Imaging (ISBI).

[23]  D. McCauley,et al.  Low-dose CT of the lungs: preliminary observations. , 1990, Radiology.

[24]  Yuan Xu,et al.  Image Restoration for Low-Dose CT via Transfer Learning and Residual Network , 2020, IEEE Access.

[25]  Zhengrong Liang,et al.  Analytical noise treatment for low-dose CT projection data by penalized weighted least-square smoothing in the K-L domain , 2002, SPIE Medical Imaging.

[26]  Quanzheng Li,et al.  A Cascaded Convolutional Nerual Network for X-ray Low-dose CT Image Denoising , 2017, ArXiv.

[27]  Steve B. Jiang,et al.  Low-dose CT reconstruction via edge-preserving total variation regularization. , 2010, Physics in medicine and biology.

[28]  Jong Chul Ye,et al.  Wavelet Domain Residual Network (WavResNet) for Low-Dose X-ray CT Reconstruction , 2017, ArXiv.

[29]  C. Zhang,et al.  CT artifact reduction via U-net CNN , 2018, Medical Imaging.

[30]  Jianhua Ma,et al.  Progressive transfer learning strategy for low-dose CT image reconstruction with limited annotated data , 2020 .

[31]  Yuxiang Xing,et al.  Slice-wise reconstruction for low-dose cone-beam CT using a deep residual convolutional neural network , 2019, Nuclear Science and Techniques.

[32]  Jiasong Wu,et al.  Improving Low-Dose CT Image Using Residual Convolutional Network , 2017, IEEE Access.

[33]  Wufan Chen,et al.  Variance analysis of x-ray CT sinograms in the presence of electronic noise background. , 2012, Medical physics.

[34]  L. Tanoue.  Computed Tomography — An Increasing Source of Radiation Exposure , 2009 .

[35]  Yuxiang Xing,et al.  Comparison of projection domain, image domain, and comprehensive deep learning for sparse-view X-ray CT image reconstruction , 2018 .

[36]  Xuanqin Mou,et al.  Low-Dose CT Image Denoising Using a Generative Adversarial Network With Wasserstein Distance and Perceptual Loss , 2017, IEEE Transactions on Medical Imaging.

[37]  Kazuo Awai,et al.  Radiation dose reduction without degradation of low-contrast detectability at abdominal multisection CT with a low-tube voltage technique: phantom study. , 2005, Radiology.

[38]  S. Kappler,et al.  Local orientation-dependent noise propagation for anisotropic denoising of CT-images , 2009, 2009 IEEE Nuclear Science Symposium Conference Record (NSS/MIC).

[39]  Zhaoying Bian,et al.  Optimizing a Parameterized Plug-and-Play ADMM for Iterative Low-Dose CT Reconstruction , 2019, IEEE Transactions on Medical Imaging.

[40]  D. Brenner,et al.  Cancer risks from diagnostic radiology. , 2008, The British journal of radiology.

[41]  Jing Wang,et al.  Adaptive nonlocal means-based regularization for statistical image reconstruction of low-dose X-ray CT , 2015, Medical Imaging.

[42]  Jeffrey A. Fessler,et al.  Statistical image reconstruction for polyenergetic X-ray computed tomography , 2002, IEEE Transactions on Medical Imaging.

[43]  Natalie N. Braun,et al.  Strategies for reducing radiation dose in CT. , 2009, Radiologic clinics of North America.

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

[45]  Jaakko Lehtinen,et al.  High-Quality Self-Supervised Deep Image Denoising , 2019, NeurIPS.