基于区域内容感知核范数的低剂量CT影像去噪

doi:10.11772/j.issn.1001-9081.2019091592

计算机应用 ›› 2020, Vol. 40 ›› Issue (4): 1177-1183.DOI: 10.11772/j.issn.1001-9081.2019091592

• 虚拟现实与多媒体计算 • 上一篇下一篇

基于区域内容感知核范数的低剂量CT影像去噪

宋芸¹, 张元科^2,3, 卢虹冰³, 邢宇翔⁴, 马建华²

1. 曲阜师范大学信息科学与工程学院, 山东日照 276800;
2. 南方医科大学生物医学工程学院, 广州 510515;
3. 空军军医大学生物医学工程系, 西安 710032;
4. 清华大学工程物理系, 北京 100084

收稿日期:2019-09-19 修回日期:2019-11-07 出版日期:2020-04-10 发布日期:2019-11-18
通讯作者: 张元科
作者简介:宋芸(1994-),女,山东烟台人,硕士研究生,主要研究方向:医学图像复原;张元科(1979-),男,山东蓬莱人,副教授,博士,主要研究方向:CT/PET成像理论与方法、医学图像复原;卢虹冰(1967-),女,陕西西安人,教授,博士,主要研究方向:CT/PET成像理论与方法;邢宇翔(1972-),女,浙江宁波人,副研究员,博士,主要研究方向:CT/PET/能谱CT成像理论与方法;马建华(1975-),男,山东枣庄人,教授,博士,主要研究方向:CT/PET/能谱CT成像理论与方法。
基金资助:
国家自然科学基金资助项目（61871383），国家重点研发计划项目（2017YFC0107400）。

Regional-content-aware nuclear norm for low-does CT image denosing

SONG Yun¹, ZHANG Yuanke^2,3, LU Hongbing³, XING Yuxiang⁴, MA Jianhua²

1. College of Information Science and Engineering, Qufu Normal University, Rizhao Shandong 276800, China;
2. School of Biomedical Engineering, Southern Medical University, Guangzhou Guangdong 510515, China;
3. Faculty of Biomedical Engineering, Air Force Medical University, Xi'an Shaanxi 710032, China;
4. Department of Engineering Physics, Tsinghua University, Beijing 100084, China

Received:2019-09-19 Revised:2019-11-07 Online:2020-04-10 Published:2019-11-18
Supported by:
This work is supported by the National Natural Science Foundation of China (61871383),the National Key Research and Development Program of China(2017YFC0107400).

摘要/Abstract

摘要： 针对传统基于核范数最小化（NNM）的低秩约束模型在低剂量CT（LDCT）影像去噪中易造成局部纹理细节丢失的问题，提出一种具有区域内容感知能力的加权NNM的LDCT影像去噪算法。首先采用基于奇异值分解（SVD）的方法估计LDCT影像中的局部噪声强度；然后采用基于局部统计特性的方法进行目标影像块匹配；最后根据影像局部噪声强度以及不同奇异值水平自适应设置核范数权重，以实现基于加权NNM的LDCT影像去噪。仿真实验结果表明，所提算法在均方根误差（RMSE）指标上较传统NNM算法、全变分最小化算法以及变换学习算法分别降低30.11%、14.38%和8.75%，在结构相似度（SSIM）指标上较上述3种算法分别提高34.24%、23.06%和11.52%。真实临床数据实验结果表明，所提算法处理结果的放射医生评价平均分为8.94，与常规剂量CT影像的评价平均分数仅差0.21，显著高于传统NNM算法、全变分最小化算法和变换学习算法的平均分。仿真及真实临床数据的实验结果表明，所提算法能够在滤除LDCT影像伪影噪声的同时，有效保持局部纹理细节信息。

关键词: 低剂量计算机断层扫描, 噪声抑制, 核范数最小化, 低秩, 区域内容感知

Abstract: The low-rank constraint model based on traditional Nuclear Norm Minimization(NNM)tends to cause local texture detail loss in the denoising of Low-Dose CT(LDCT)image. To tackle this issue,a regional-content-aware weighted NNM algorithm was proposed for LDCT image denoising. Firstly,a Singular Value Decomposition(SVD)based method was proposed to estimate the local noise intensity in LDCT image. Then,the target image block matching was performed based on the local statistical characteristics. Finally,the weights of the nuclear norms were adaptively set based on both the local noise intensity of the image and the different singular value levels,and the weighted NNM based LDCT image denoising was realized. The simulation results illustrated that the proposed algorithm decreased the Root Mean Square Error(RMSE)index by 30. 11%,14. 38% and 8. 75% respectively compared with the traditional NNM,total variation minimization and transform learning algorithms,and improved the Structural SIMilarity(SSIM)index by 34. 24%,23. 06% and 11. 52% respectively compared with the above three algorithms. The experimental results on real clinical data illustrated that the mean value of the radiologists' scores of the results obtained by the proposed algorithm was 8. 94,which is only 0. 21 lower than that of the corresponding full dose CT images,and was significantly higher than those of the traditional NNM,total variation minimization and transform learning algorithms. The simulation and clinical experimental results indicate that the proposed algorithm can effectively reduce the artifact noise while preserving the texture detail information in LDCT images.

Key words: Low-Does Computed Tomography (LDCT), noise reduction, Nuclear Norm Minimization (NNM), low rank, regional content awareness

中图分类号:

TP391.41

宋芸, 张元科, 卢虹冰, 邢宇翔, 马建华. 基于区域内容感知核范数的低剂量CT影像去噪[J]. 计算机应用, 2020, 40(4): 1177-1183.

SONG Yun, ZHANG Yuanke, LU Hongbing, XING Yuxiang, MA Jianhua. Regional-content-aware nuclear norm for low-does CT image denosing[J]. Journal of Computer Applications, 2020, 40(4): 1177-1183.

参考文献

[1] BRENNER D J,HALL E J. Cancer risks from CT scans:now we have data,what next?[J]. Radiology,2012,265(2):330-331.
[2] MOSER J B,SHEARD S L,EDYEAN S,et al. Radiation dose-reduction strategies in thoracic CT[J]. Clinical Radiology,2017, 72(5):407-420.
[3] BRENNER D J,HALL E J. Computed tomography-an increasing source of radiation exposure[J]. New England Journal of Medicine, 2007,357(22):2277-2284.
[4] PAUL N S,BLOBEL J,PREZELJ E,et al. The reduction of image noise and streak artifact in the thoracic inlet during low dose and ultra-low dose thoracic CT[J]. Physics in Medicine and Biology, 2010,55(5):No. 1363.
[5] LAMBIN P,LEIJENAAR R T H,DEIST T M,et al. Radiomics:the bridge between medical imaging and personalized medicine[J]. Nature Reviews Clinical Oncology,2017,14(12):749-762.
[6] HU Z,GAO J,ZHANG N,et al. An improved statistical iterative algorithm for sparse-view and limited-angle CT image reconstruction[J]. Scientific Reports,2017,7(1):No. 10747.
[7] 杜毓菁, 王荣福. 全景PET/CT的研究进展及轴向视场的新突破[J]. CT理论与应用研究,2018,27(5):675-682. (DU Y J, WANG R F. Research progress of PET/CT and new breakthrough in axial field of view[J]. CT Theory and Applications,2018,27(5):675-682.)
[8] SHI C T,CHANG S J,LIU Y L,et al. Noise reduction of low-dose computed tomography using the multi-resolution total variation minimization algorithm[C]//Proceedings of the Medical Imaging 2013:Physics of Medical Imaging,SPIE 8668. Bellingham,WA:SPIE,2013:No. 86682H.
[9] LIU W,CAO S,CHEN Y. Seismic time-frequency analysis via empirical wavelet transform[J]. IEEE Geoscience and Remote Sensing Letters,2016,13(1):28-32.
[10] DIWAKAR M,KUMAR M. Edge preservation based CT image denoising using Wiener filtering and thresholding in wavelet domain[C]//Proceedings of the 4th International Conference on Parallel,Distributed and Grid Computing. Piscataway:IEEE, 2017:332-336.
[11] ZVOLANEK K,MA R,ZHOU C,et al. Still equivalent for dose calculation in the Monte Carlo era? A comparison of free breathing and average intensity projection CT datasets for lung SBRT using three generations of dose calculation algorithms[J]. Medical Physics,2017,44(5):1939-1947.
[12] ABUBAKAR A,ZHAO X J,LI S T,et al. A block-matching and 3-D filtering algorithm for Gaussian noise in DoFP polarization images[J]. IEEE Sensors Journal,2018,18(18):7429-7435.
[13] 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.
[14] 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):No. 2103.
[15] 蒋慧琴, 徐玉风, 马岭, 等. 一种自适应低剂量CT图像质量改善算法[J]. 郑州大学学报(工学版),2018,39(4):75-80. (JIANG H Q,XU Y F,MA L,et al. An adaptive quality improved algorithm in low dose CT images[J]. Journal of Zhengzhou University (Engineering Science),2018,39(4):75-80.)
[16] HASAN M,EL-SAKKA M R. Improved BM3D image denoising using SSIM-optimized Wiener filter[J]. EURASIP Journal on Image and Video Processing,2018,2018(1):No. 25.
[17] ALKINANI M H,EL-SAKKA M R. Patch-based models and algorithms for image denoising:a comparative review between patch-based images denoising methods for additive noise reduction[J]. EURASIP Journal on Image and Video Processing,2017, 2017(1):No. 58.
[18] CHEN W,SHAO Y,WANG Y,et al. A novel total variation model for low-dose CT image denoising[J]. IEEE Access,2018, 6:78892-78903.
[19] 何琳, 张权, 上官宏, 等. 低剂量CT图像的自适应广义总变分降噪算法[J]. 计算机应用,2016,36(1):243-247.(HE L, ZHANG Q,SHANGGUAN H,et al. Adaptive total generalized variation denoising algorithm for low-dose CT images[J]. Journal of Computer Applications,2016,36(1):243-247.)
[20] 王娜, 张权, 刘祎, 等. 基于可变阶变分模型的医用低剂量CT图像去噪[J]. 北京航空航天大学学报,2019,45(9):1757-1764.(WANG N,ZHANG Q,LIU Y,et al. Medical low-dose CT image denoising based on variable order variational model[J]. Journal of Beijing University of Aeronautics and Astronautics, 2019,45(9):1757-1764.)
[21] KANG E,CHANG W,YOO J,et al. Deep convolutional framelet denosing for low-dose CT via wavelet residual network[J]. IEEE Transactions on Medical Imaging,2018,37(6):1358-1369.
[22] PENG Z, WANG G. Study on optimal selection of wavelet vanishing moments for ECG denoising[J]. Scientific Reports, 2017,7(1):No. 4564.
[23] YANG D,HU S,LIU S,et al. Multi-focus image fusion based on block matching in 3D transform domain[J]. Journal of Systems Engineering and Electronics,2018,29(2):415-428.
[24] ZHANG Y,RONG J,LU H,et al. Low-dose lung CT image restoration using adaptive prior features from full-dose training database[J]. IEEE Transactions on Medical Imaging,2017,36(12):2510-2523.
[25] ZHANG Y,LU H,RONG J,et al. Adaptive non-local means on local principle neighborhood for noise artifacts reduction of low-dose CT images[J]. Medical Physics,2017,44(9):e230-e241.
[26] ELAD M, AHARON M. Image denoising via sparse and redundant representations over learned dictionaries[J]. IEEE Transactions on Image Processing,2006,15(12):3736-3745.
[27] 李中源, 李光, 孙翌, 等. 一种基于全局字典学习的小动物低剂量计算机断层扫描降噪方法[J]. 生物医学工程学杂志,2016, 33(2):279-286.(LI Z Y,LI G,SUN Y,et al. A denoising method for low-does small-animal computed tomography image based on globe dictionary learning[J]. Journal of Biomedical Engineering,2016,33(2):279-286.)
[28] 朱永成, 陈阳, 罗立民, 等. 基于字典学习的低剂量X-ray CT图像去噪[J]. 东南大学学报(自然科学版),2012,42(5):864-868.(ZHU Y C,CHEN Y,LUO L M,et al. Dictionary learning based denoising of low-dose X-ray CT image[J]. Journal of Southeast University(Natural Science Edition),2012,42(5):864-868.)
[29] WEN B, RAVISHANKAR S, BRESLER Y. Structured overcomplete sparsifying transform learning with convergence guarantees and applications[J]. International Journal of Computer Vision,2015,114(2/3):137-167.
[30] YOON H,KIM K S,KIM D,et al. Motion adaptive patch-based low-rank approach for compressed sensing cardiac cine MRI[J]. IEEE Transactions on Medical Imaging,2014,33(11):2069-2085.
[31] LIU S,ZHANG T,LI H,et al. Medical image fusion based on nuclear norm minimization[J]. International Journal of Imaging Systems and Technology,2015,25(4):310-316.
[32] CAI J,CANDÈS E J,SHEN Z. A singular value thresholding algorithm for matrix completion[J]. SIAM Journal on Optimization,2010,20(4):1956-1982.
[33] JOLLIFE I T. Principal Component Analysis[M]. New York:Springer,1986:135-137.
[34] LU C, ZHU C, XU C, et al. Generalized singular value thresholding[C]//Proceedings of the 29th AAAI Conference on Artificial Intelligence. Palo Alto,CA:AAAI Press,2015:1805-1811.
[35] FROSIO I,KAUTZ J. Statistical nearest neighbors for image denoising[J]. IEEE Transactions on Image Processing,2018,28(2):723-738.
[36] MA J,LIANG Z,FAN Y,et al. Variance analysis of X-ray CT sinograms in the presence of electronic noise background[J]. Medical Physics,2012,39(7 Pt 1):4051-4065.
[37] ZHANG H,HAN H,LIANG Z,et al. Extracting information from previous full-dose CT scan for knowledge-based Bayesian reconstruction of current low-dose CT images[J]. IEEE Transactions on Medical Imaging,2016,35(3):860-870.
[38] WANG Z,BOVIK A C,SHEIKH H R,et al. Image quality assessment:from error visibility to structural similarity[J]. IEEE Transactions on Image Processing,2004,13(4):600-612.

基于区域内容感知核范数的低剂量CT影像去噪

Regional-content-aware nuclear norm for low-does CT image denosing

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	严海升, 马新强. 基于径向基函数的多目标回归特征构建算法[J]. 计算机应用, 2021, 41(8): 2219-2224.
[2]	王心, 朱浩华, 刘光灿. 卷积鲁棒主成分分析[J]. 计算机应用, 2021, 41(5): 1314-1318.
[3]	李杏峰, 黄玉清, 任珍文, 李毅红. 基于自适应邻域的鲁棒多视图聚类算法[J]. 计算机应用, 2021, 41(4): 1093-1099.
[4]	蒋丽, 黄仕建, 严文娟. 基于低秩行为信息和多尺度卷积神经网络的人体行为识别方法[J]. 计算机应用, 2021, 41(3): 721-726.
[5]	李杏峰, 黄玉清, 任珍文. 联合低秩稀疏的多核子空间聚类算法[J]. 计算机应用, 2020, 40(6): 1648-1653.
[6]	张丽, 孔旭, 孙忠贵. 基于非局部自相似性和低秩矩阵逼近的补全算法[J]. 计算机应用, 2020, 40(11): 3327-3331.
[7]	陈利霞, 刘俊丽, 王学文. 结合加权Schatten-p范数与3D全变分的前景检测[J]. 计算机应用, 2019, 39(4): 1170-1175.
[8]	张雯雯, 韩裕生. 基于非局部自相似性的低秩稀疏图像去噪[J]. 计算机应用, 2018, 38(9): 2696-2700.
[9]	杨美姣, 刘惊雷. 基于Nyström方法的偏好特征提取[J]. 计算机应用, 2018, 38(9): 2515-2522.
[10]	张乐园, 李佳烨, 李鹏清. 低秩约束的非线性属性选择算法[J]. 计算机应用, 2018, 38(12): 3444-3449.
[11]	高净植, 刘祎, 白旭, 张权, 桂志国. 平稳小波域深度残差CNN用于低剂量CT图像估计[J]. 计算机应用, 2018, 38(12): 3584-3590.
[12]	孙少超. 基于带权核范数最小化和混合高斯模型的去噪模型[J]. 计算机应用, 2017, 37(5): 1471-1474.
[13]	徐文超, 王光艳, 陈雷. 改进的变步长最小均方误差电子耳蜗语音增强算法[J]. 计算机应用, 2017, 37(4): 1212-1216.
[14]	甘玲, 左永强. 基于快速低秩编码与局部约束的图像分类算法[J]. 计算机应用, 2017, 37(10): 2912-2915.
[15]	张鹏程, 张权, 张芳, 陈燕, 韩建宁, 郝慧艳, 桂志国. 结合改进扩散和双边滤波的低剂量CT重建[J]. 计算机应用, 2016, 36(4): 1100-1105.