A dual-domain deep learning-based reconstruction method for fully 3D sparse data helical CT

Helical CT has been widely used in clinical diagnosis. In this work, we focus on a new prototype of helical CT which equipped with sparsely spaced multidetector and multi-slit collimator (MSC) in the axis direction. This kind systems can not only lower radiation dose, suppress scattering by MSC, but also cut down the manufacturing cost of detector. The major problem to overcome for such a system is the insufficient data for reconstruction. Hence, we propose a deep learning-based function optimization method for this ill-posed inverse problem. By incorporating Radon inverse operator and disentangling each slice, we significantly simplify the complexity of our network for 3D reconstruction. The whole network is composed of three subnetworks. First, a convolutional neural network (CNN) in projection domain is constructed to estimate missing projection data and convert helical projection data to 2D fan-beam projection data. Then, an analytical linear operator is followed to transfer the data from projection domain to image domain. Finally, an additional CNN in image domain is added for further image refinement. These three steps work as an entirety and can be trained end to end. The overall network is trained on a simulated CT dataset from 8 patients of American Association of Physicists in Medicine (AAPM) Low-Dose CT Grand Challenge. We evaluate the trained network on both simulated datasets and clinical datasets. From extensive experimental studies, very encouraging results are obtained according to both visual examination and quantitative evaluation. These results demonstrated the effectiveness of our method and its potential for clinical usage. The proposed method provides us a new solution for a fully 3D ill-posed problem.

[1]  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.

[2]  Alexander Katsevich,et al.  Fast Exact/Quasi-Exact FBP Algorithms for Triple-Source Helical Cone-Beam CT , 2010, IEEE Transactions on Medical Imaging.

[3]  Alexander Katsevich,et al.  Theoretically Exact Filtered Backprojection-Type Inversion Algorithm for Spiral CT , 2002, SIAM J. Appl. Math..

[4]  Jeffrey A. Fessler,et al.  SPULTRA: Low-Dose CT Image Reconstruction With Joint Statistical and Learned Image Models , 2018, IEEE Transactions on Medical Imaging.

[5]  Ke Li,et al.  Statistical model based iterative reconstruction (MBIR) in clinical CT systems: experimental assessment of noise performance. , 2014, Medical physics.

[6]  Bruce R. Whiting,et al.  Signal statistics in x-ray computed tomography , 2002, SPIE Medical Imaging.

[7]  Jeffrey A. Fessler,et al.  Deep BCD-Net Using Identical Encoding-Decoding CNN Structures for Iterative Image Recovery , 2018, 2018 IEEE 13th Image, Video, and Multidimensional Signal Processing Workshop (IVMSP).

[8]  Hu Chen,et al.  LEARN: Learned Experts’ Assessment-Based Reconstruction Network for Sparse-Data CT , 2017, IEEE Transactions on Medical Imaging.

[9]  Yuxiang Xing,et al.  Improve angular resolution for sparse-view CT with residual convolutional neural network , 2018, Medical Imaging.

[10]  Ziheng Li,et al.  Promising Generative Adversarial Network Based Sinogram Inpainting Method for Ultra-Limited-Angle Computed Tomography Imaging , 2019, Sensors.

[11]  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).

[12]  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..

[13]  Ge Wang,et al.  Dual-dictionary learning-based iterative image reconstruction for spectral computed tomography application , 2012, Physics in medicine and biology.

[14]  Mannudeep K. Kalra,et al.  Low-Dose CT with a Residual Encoder-Decoder Convolutional Neural Network (RED-CNN) , 2017, ArXiv.

[15]  Quanzheng Li,et al.  Computationally Efficient Cascaded Training for Deep Unrolled Network in CT Imaging , 2018, Medical physics.

[16]  Michael Unser,et al.  Deep Convolutional Neural Network for Inverse Problems in Imaging , 2016, IEEE Transactions on Image Processing.

[17]  Daniel K Sodickson,et al.  SparseCT: System concept and design of multislit collimators. , 2019, Medical physics.

[18]  Jeffrey A. Fessler,et al.  Convolutional Dictionary Learning: Acceleration and Convergence , 2017, IEEE Transactions on Image Processing.

[19]  Jong Chul Ye,et al.  A deep convolutional neural network using directional wavelets for low‐dose X‐ray CT reconstruction , 2016, Medical physics.

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

[21]  H. Kudo,et al.  Helical-scan computed tomography using cone-beam projections , 1991 .

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

[23]  Jeffrey A. Fessler,et al.  Adaptive sparse modeling and shifted-poisson likelihood based approach for low-dosect image reconstruction , 2017, 2017 IEEE 27th International Workshop on Machine Learning for Signal Processing (MLSP).

[24]  Ke Li,et al.  Learning to Reconstruct Computed Tomography Images Directly From Sinogram Data Under A Variety of Data Acquisition Conditions , 2019, IEEE Transactions on Medical Imaging.

[25]  Bruce R. Rosen,et al.  Image reconstruction by domain-transform manifold learning , 2017, Nature.

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

[27]  L. Rudin,et al.  Nonlinear total variation based noise removal algorithms , 1992 .

[28]  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.

[29]  Andreas K. Maier,et al.  Learning with Known Operators reduces Maximum Training Error Bounds , 2019, Nature Machine Intelligence.

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

[31]  Huazhong Shu,et al.  Domain Progressive 3D Residual Convolution Network to Improve Low-Dose CT Imaging , 2019, IEEE Transactions on Medical Imaging.

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

[33]  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.

[34]  Ricardo Otazo,et al.  Image reconstruction for interrupted-beam x-ray CT on diagnostic clinical scanners , 2019, Physics in medicine and biology.

[35]  Rolf Clackdoyle,et al.  Cone-beam reconstruction using the backprojection of locally filtered projections , 2005, IEEE Transactions on Medical Imaging.

[36]  Mathias Unberath,et al.  Deep Learning Computed Tomography: Learning Projection-Domain Weights From Image Domain in Limited Angle Problems , 2018, IEEE Transactions on Medical Imaging.

[37]  Xiaochuan Pan,et al.  Exact image reconstruction on PI-lines from minimum data in helical cone-beam CT. , 2004, Physics in medicine and biology.