北京生物医学工程

核范数最小化在低剂量CT去噪中的研究

Denoising of low-dose CT images based on the nuclear norm minimization method

作者：王爽

单位：北京交通大学计算机与信息技术学院（北京 100044）

分类号：R318.04

出版年·卷·期（页码）：2020·39·5（484-490）

摘要：

目的加权核范数最小化（weighted nuclear norm minimization, WNNM）算法是一种低秩矩阵逼近去噪方法，目前主要应用于自然图像去噪。本文拟探究WNNM方法在低剂量CT图像去噪中应用价值。方法将WNNM算法与三维块匹配滤波（block-matching and 3d filtering, BM3D）算法相结合，提出BM3D-WNNM图像后处理方法。使用该方法对Shepp-Logan仿真低剂量CT图像做去噪处理，并将其结果与BM3D算法、WNNM算法以及基于直觉模糊熵和加权方差的各向异性扩散算法(anisotropic diffusion algorithm based on intuitionistic fuzzy entropy and weighted variance, IFE-WV-AD)的结果进行对比分析，定量比较4种算法对Shepp-Logan仿真低剂量CT图像去噪结果的峰值信噪比（peak signal to noise ratio, PSNR）和结构相似性（structural similarity, SSIM）。结果 BM3D-WNNM方法处理后的图像在视觉观察和定量分析中均优于另外3种方法，该方法处理后的图像中肉眼可分辨的斑点状噪声和条形伪影少于另外3种算法处理后的图像，且 PSNR和SSIM均大于另外3种算法的结果。结论 BM3D-WNNM算法可以有效去除Shepp-Logan仿真低剂量CT图像中的噪声，在低剂量CT图像去噪中具有应用价值。

Objective The weighted nuclear norm minimization (WNNM) algorithm is a low rank matrix approximation denoising method, which is currently mainly applied to natural image denoising. This paper intends to explore the application value of WNNM method in low-dose CT images denoising. Methods Combining the WNNM algorithm with a block-matching and 3d filtering (BM3D) algorithm ， an image post-processing method （BM3D-WNNM algorithm）is proposed. This method is used to denoise the Shepp-Logan simulated low-dose CT images.The results are compared with the results of BM3D algorithm, WNNM algorithm and anisotropic diffusion algorithm based on intuitionistic fuzzy entropy and weighted variance (IFE-WV-AD)，and quantitatively compared the peak signal to noise ratio (PSNR) and structural similarity (SSIM) of the Shepp-Logan simulated low-dose CT images denoised by four algorithms. Results The images processed by BM3D-WNNM algorithm are superior to the other three methods both in visual observation and quantitative analysis. The visually distinguishable speckle noise and stripe artifacts in images processed by the proposed method are less than those processed by the other three algorithms, the PSNR and SSIM values of the images processed by this method are larger than the results of the other three algorithms. Conclusions The BM3D-WNNM algorithm can effectively remove the noise in Shepp-Logan simulation low-dose CT images, and has application value in low-dose CT image denoising.

参考文献：

[1] Brenner DJ , Elliston CD , Hall E J , et al. Estimated risk of radiation –induced fatal cancer from pediatric CT[J]. American Journal of Roentgenology, 2001, 176(2):289-296.

[2] Naidich D. Low-dose CT of the lungs: preliminary observations[J]. Radiology, 1990, 175(3):729-31.

[3] 崔学英. 低剂量CT的投影域去噪算法和后处理方法研究[D].太原：中北大学, 2015.

Cui XY. Research on sinogram noise reduction method and post-processing approaches for low-dose CT[D]. Taiyuan:North University of China, 2015.

[4] Karimi D , Deman P , Ward R , et al. A sinogram denoising algorithm for low-dose computed tomography[J]. BMC Medical Imaging, 2016, 16(1):11.

[5] Gu S, Zhang L, Zuo W, et al. Weighted nuclear norm minimization with application to image denoising[C]// Computer Vision and Pattern Recognition. IEEE, 2014:2862-2869.

[6] Buchanan AM, Fitzgibbon AW. Buchanan A M , Fitzgibbon A W . Damped Newton Algorithms for Matrix Factorization with Missing Data[C]// IEEE Computer Society Conference on Computer Vision & Pattern Recognition. IEEE, 2005:316-322.

[7] Cai JF , Candès EJ, Shen ZW . A singular value thresholding algorithm for matrix completion[J]. SIAM Journal on Optimization, 2010, 20(4):1956-1982.

[8] Donoho DL, Gavish M, Montanari A. The phase transition of matrix recovery from Gaussian measurements matches the minimax MSE of matrix denoising[J]. Proceedings of the National Academy of Sciences, 2013, 110(21):8405-8410.

[9] Gao H, Cai J F, Shen Z, et al. Robust principal component analysis-based four-dimensional computed tomography[J]. Physics in Medicine and Biology, 2011, 56(11):3181-3198.

[10] Cai JF, Jia X, Gao H, et al. Cine cone beam CT reconstruction using low-rank matrix factorization: algorithm and a proof-of-princple study[J]. IEEE Transactions on Medical Imaging, 2014, 33(8):1581-1591.

[11]马海英,宣士斌,向顺灵.低秩矩阵在CT图像重建中的应用[J].广西民族大学学报(自然科学版),2016,22(3):86-92.

Ma HY, Xuan SB, Xiang SL. Application of low rank matrix in ct image reconstruction[J]. Journal of Guangxi University for Nationalities (Natural Science Edition) ,2016,22(3):86-92.

[12] Dabov K, Foi A, Katkovnik V, et al. Image denoising by sparse 3-D transform-domain collaborative filtering[J]. IEEE Transactions on Image Processing, 2007, 16(8):2080-2095.

[13] Diwakar M, Kumar M. A review on CT image noise and its denoising[J]. Biomedical Signal Processing & Control, 2018, 42:73-88.

[14] Zamyatin A, Krylov R, Shi B, et al. Adaptive multi-scale total variation minimization filter for low dose CT imaging [C]// SPIE. Physics of Medical Imaging.New York:Society of Photo-optical Instrumentation Engineers,2014:903426.

[15] Chen Y , Shi L , Feng Q , et al. Artifact Suppressed Dictionary Learning for Low-Dose CT Image Processing[J]. IEEE Transactions on Medical Imaging, 2014, 33(12):2271-2292.

[16] Bai T, Mou X, Xu Q, et al. Low-dose CT reconstruction based on multiscale dictionary[C]// SPIE. Physics of Medical Imaging.New York:Society of Photo-optical Instrumentation Engineers,2013.

[17] Shi L, Chen Y,Shu H, et al. Low-dose CT image processing using artifact suppressed dictionary learning[C]. 2014 IEEE 11th International Symposium on Biomedical Imaging. Beijing:ISBI,2014:1127-1130.

[18] 石路遥. 基于字典学习的低剂量CT图像处理方法[D].南京：东南大学,2015.

Shi LY. Dictionary learning based low-dose CT image processing methods[D]. Nanjing：Southeast University, 2015.

[19] Chen Y, Liu J, Hu Y, et al. Discriminative feature representation: an effective postprocessing solution to low dose CT imaging[J]. Physics in Medicine and Biology, 2017, 62(6):2103-2131.

[20] Xu Q, Yu H, Mou X, et al. Low-dose X-ray CT reconstruction via dictionary learning[J]. IEEE Transactions on Medical Imaging, 2012, 31(9):1682-1697.

[21] Chen Y, Yin X, Shi L, et al. Improving abdomen tumor low-dose CT images using a fast dictionary learning based processing[J]. Physics in Medicine and Biology, 2013, 58(16):5803-5820.

[22] 张旭. 低剂量CT图像的后处理算法[D].太原：中北大学,2015.

Zhang Xu, Low dose CT image post-processing algorithm[D]. Taiyuan：North University of China, 2015.

[23] Chen H, Zhang Y, Zhang W, et al. Low-dose CT via convolutional neural network [J]. Biomedical Optics Express, 2017, 8(2): 679.

[24] Chen H , Zhang Y , Kalra M K , et al. Low-dose CT with a residual encoder-decoder convolutional neural network [J]. IEEE Transactions on Medical Imaging, 2017,36(12): 2524-2535.

[25] Yang W, Zhang H, Yang J, et al. Improving Low-Dose CT Image Using Residual Convolutional Network[J]. IEEE Access, 2017, 5:24698-24705.

[26] Wu D , Kim K , Fakhri GE , et al. A cascaded convolutional neural network for X-ray low-dose ct image denoising[J/OL]. (2017-08-28). arXiv:1705.04267. https://arxiv.org/abs/1705.04267.

[27] Wolterink JM , Leiner T , Viergever M A , et al. Generative adversarial networks for noise reduction in low-dose CT[J]. IEEE Transactions on Medical Imaging, 2017, 36(12):2536-2545.

[28] Yang Q , Yan P , Zhang Y , et al. Low dose CT image denoising using a generative adversarial network with wasserstein distance and perceptual loss[J]. IEEE Transactions on Medical Imaging, 2018, 37(6): 1348-1357.

[29] Yi X , Babyn P . Sharpness-aware low-dose CT denoising using conditional generative adversarial network[J]. Journal of Digital Imaging, 2018,31: 655–669.

[30] Chun Jiao, Dongming Wang, Hongbing Lu, et al. Multiscale noise reduction on low-dose CT Sinogram by stationary wavelet transform[C]// 2008 IEEE Nuclear Science Symposium Conference Record.Dresden, Germany：IEEE， 2008： 5339-5344.

[31] 刘进. 特征稀疏表示的低剂量CT成像方法研究[D].南京：东南大学,2018.

Liu J, Feature sparse representation based low dose CT imaging[D]. Nanjing：Southeast University, 2018.

[32] Chen Y, Liu J, Hu Y, et al. Discriminative feature representation: an effective postprocessing solution to low dose CT imaging[J]. Physics in Medicine and Biology, 2017, 62(6):2103-2131.

[33] Zeng D, Huang J, Bian Z, et al. A simple low-dose X-ray CT simulation from high-dose scan[J]. IEEE Transactions on Nuclear Science, 2015, 62(5):2226-2233.

[34] Sheikh H R, Bovik A C. Image information and visual quality[J]. IEEE Transactions on Image Processing, 2006, 15(2):430-444.

[35] Wang Z, Bovik A, Sheich HR, et al. Image quality assessment: from error visibility to structural similarity[J].IEEE transactions on image processing : a publication of the IEEE Signal Processing Society, 2004, 13(4):600-612.

服务与反馈：

【文章下载】【加入收藏】

提示：您还未登录，请登录！点此登录