Low-dose CT image denoising using residual convolutional network with fractional TV loss

Abstract In this work, we propose a Fractional-order Residual Convolutional Neural Network (FRCNN) for Low-Dose CT (LDCT) denoising. As increasing the dose of radiation is harmful to the patient, how to trade off between reducing the radiation dose and improving the quality of the CT image has become a challenging problem. To this end, this paper proposes a new approach for LDCT image denoising using Convolutional Neural Network (CNN) with Fractional-order Total Variation (FTV) loss, as well as residual learning. Firstly, this paper introduced the FTV loss function for structural details enhancement. Secondly, skip connections were added to optimize the network. Thirdly, extensive experimental analysis was used to evaluate the capacity of this method in suppressing noise and preserving detailed information. The FTV loss can retain essential structural details while suppressing noise, generating high-quality CT images ready for interpretation by radiologists. Compared with state-of-the-art methods, our method obtained better results visually and numerically, especially in structural details preservation. These promising results will significantly improve the usability of LDCT images.

[1]  Roland R. Draxler,et al.  Root mean square error (RMSE) or mean absolute error (MAE) , 2014 .

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

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

[4]  W. Kalender,et al.  Generalized multi-dimensional adaptive filtering for conventional and spiral single-slice, multi-slice, and cone-beam CT. , 2001, Medical physics.

[5]  R. Weissleder,et al.  Block matching 3D random noise filtering for absorption optical projection tomography , 2010, Physics in medicine and biology.

[6]  J. Coatrieux,et al.  Improving abdomen tumor low-dose CT images using a fast dictionary learning based processing , 2013, Physics in medicine and biology.

[7]  Guo-Cheng Wu,et al.  New variable-order fractional chaotic systems for fast image encryption. , 2019, Chaos.

[8]  Weixing Wang,et al.  Fractional differential approach to detecting textural features of digital image and its fractional differential filter implementation , 2008, Science in China Series F: Information Sciences.

[9]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[10]  Xiaoou Tang,et al.  Image Super-Resolution Using Deep Convolutional Networks , 2014, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[11]  Zhang Yi,et al.  A Fractional-Order Variational Framework for Retinex: Fractional-Order Partial Differential Equation-Based Formulation for Multi-Scale Nonlocal Contrast Enhancement with Texture Preserving , 2018, IEEE Transactions on Image Processing.

[12]  M. Osanai,et al.  Photon starvation artifacts of X-ray CT: their true cause and a solution , 2012, Radiological Physics and Technology.

[13]  L. Xing,et al.  Iterative image reconstruction for CBCT using edge-preserving prior. , 2008, Medical physics.

[14]  Jürgen Schmidhuber,et al.  Training Very Deep Networks , 2015, NIPS.

[15]  Karen O. Egiazarian,et al.  Image restoration by sparse 3D transform-domain collaborative filtering , 2008, Electronic Imaging.

[16]  Zhaoying Bian,et al.  A Simple Low-Dose X-Ray CT Simulation From High-Dose Scan , 2015, IEEE Transactions on Nuclear Science.

[17]  Hu Chen,et al.  Low-dose CT via convolutional neural network. , 2017, Biomedical optics express.

[18]  Zhengrong Liang,et al.  Detector response restoration in image reconstruction of high resolution positron emission tomography , 1994, IEEE Trans. Medical Imaging.

[19]  Vito Pascazio,et al.  Statistical regularization in linearized microwave imaging through MRF-based MAP estimation: hyperparameter estimation and image computation , 2003, IEEE Trans. Image Process..

[20]  Guo-Cheng Wu,et al.  Discrete tempered fractional calculus for new chaotic systems with short memory and image encryption , 2020 .

[21]  Steven L. Fritz,et al.  A 57cm X-ray Image Intensifier Digital Radiography System , 1985 .

[22]  Lei Xing,et al.  SU‐FF‐I‐44: Iterative Image Reconstruction for CBCT Using Edge‐Preserving Prior , 2009 .

[23]  M. Elad,et al.  $rm K$-SVD: An Algorithm for Designing Overcomplete Dictionaries for Sparse Representation , 2006, IEEE Transactions on Signal Processing.

[24]  Ji-Huan He,et al.  Two-scale mathematics and fractional calculus for thermodynamics , 2019, Thermal Science.

[25]  Jian Huang,et al.  Kernel machine-based one-parameter regularized Fisher discriminant method for face recognition , 2005, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[26]  Cynthia M. McCollough,et al.  Projection space denoising with bilateral filtering and CT noise modeling for dose reduction in CT. , 2009, Medical physics.

[27]  Yi-Fei Pu,et al.  Fractional Differential Mask: A Fractional Differential-Based Approach for Multiscale Texture Enhancement , 2010, IEEE Transactions on Image Processing.

[28]  K. B. Oldham,et al.  The Fractional Calculus: Theory and Applications of Differentiation and Integration to Arbitrary Order , 1974 .

[29]  Lei Zhang,et al.  Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising , 2016, IEEE Transactions on Image Processing.

[30]  Ji-Huan He,et al.  New promises and future challenges of fractal calculus: From two-scale thermodynamics to fractal variational principle , 2020 .

[31]  YangQuan Chen,et al.  Fractional-order TV-L2 model for image denoising , 2013 .

[32]  Yuquan Chen,et al.  Lyapunov functions for nabla discrete fractional order systems. , 2019, ISA transactions.

[33]  Armando Manduca,et al.  Adaptive nonlocal means filtering based on local noise level for CT denoising. , 2013, Medical physics.

[34]  K. Lange,et al.  EM reconstruction algorithms for emission and transmission tomography. , 1984, Journal of computer assisted tomography.

[35]  Jing Wang,et al.  Penalized weighted least-squares approach to sinogram noise reduction and image reconstruction for low-dose X-ray computed tomography , 2006, IEEE Transactions on Medical Imaging.

[36]  J. Remy,et al.  Dose reduction in spiral CT angiography of thoracic outlet syndrome by anatomically adapted tube current modulation , 2001, European Radiology.