结合感知损失与双重对抗网络的低剂量CT图像去噪

doi:10.19678/j.issn.1000-3428.0063806

摘要/Abstract

摘要： 低剂量计算机断层扫描（LDCT）成像技术在医学诊断中得到广泛应用，但其斑纹噪声和非平稳条纹伪影复杂，目前多数算法仅依靠推断条件后验概率来实现图像去噪，无法应对LDCT图像噪声复杂、数据量少、先验知识缺乏的问题。提出一种结合感知损失的双重对抗网络去噪算法，以实现LDCT图像复原。该算法包含一个去噪器和一个生成器，分别从图像去噪和噪声生成2个角度来建模干净-噪声图像对的联合分布，通过联合学习使得去噪器和生成器相互指导，从而充分学习数据中的噪声信息和清晰图像信息，且学习到的去噪器可以直接用于LDCT图像修复。考虑到通过感知损失学习语义特征差异可以使去噪结果保留更多的细节和边缘信息，提出一种掩膜自监督方法，针对CT图像域训练一个语义特征提取网络用于计算感知损失。实验结果表明，与BM3D、RED-CNN、WGAN-VGG等主流去噪算法相比，该算法可以有效抑制噪声并去除伪影，最大程度地保留边缘轮廓和纹理细节，产生更符合人眼视觉特性的去噪效果，与当下LDCT图像去噪性能较好的SACNN算法相比，所提算法的PSNR和SSIM指标分别提升1.26 dB和1.8%。

关键词: 双重对抗网络, 低剂量CT图像, 噪声生成, 自监督, 感知损失

Abstract: Low-Dose Computer Tomography(LDCT) imaging technology has been widely used for medical diagnosis, but its speckle noise and nonstationary fringe artifacts are complex.Currently, most algorithms only rely on inferential conditional posterior probability to achieve image denoising, which does not deal with the complex noise, small amount of data, and lack of prior knowledge of LDCT images.A dual adversarial network denoising algorithm combined with perceptual loss is proposed to achieve LDCT image restoration.The algorithm includes a denoiser and a generator, which jointly models the joint distribution of clean-noise image pairs from the perspectives of image denoising and noise generation, respectively. By joint learning, the denoiser and generator can guide each other to fully learn the noise information and clear image information in the data, and the learned denoiser can be directly used for LDCT image repairs.Considering that more details and edge information can be retained in denoising results by learning semantic feature differences through perceptual loss, a mask self-supervised method is proposed to train a semantic feature extraction network for the CT image domain to calculate the perceptual loss.The experimental results show that compared with mainstream denoising algorithms such as BM3D, RED-CNN, WGAN-VGG, this algorithm can effectively suppress noise and remove artifacts, maximally retain edge contour and texture details, and induce a denoising effect more consistent with human visual characteristics. Compared with the current SACNN algorithm with improved denoising performance for LDCT images, the Peak Signal-to-Noise Ratio(PSNR) and Structural Similarity(SSIM) indicators of the proposed algorithm are improved by 1.26 dB and 1.8%, respectively.

Key words: dual adversarial network, Low-Dose CT(LDCT) image, noise generation, self supervision, perceptual loss

中图分类号:

TP391

熊景琦, 桑庆兵, 胡聪. 结合感知损失与双重对抗网络的低剂量CT图像去噪[J]. 计算机工程, 2023, 49(2): 213-221,230.

XIONG Jingqi, SANG Qingbing, HU Cong. Low-Dose CT Image Denoising Combining Perceptual Loss and Dual Adversarial Network[J]. Computer Engineering, 2023, 49(2): 213-221,230.

http://www.ecice06.com/CN/Y2023/V49/I2/213

图/表 9

20230216181601

20230216181605

20230216181609

20230216181612

20230216181616

20230216181619

20230216181622

20230216181625

20230216181628

参考文献

[1] BRENNER D J, HALL E J.Computed tomography-an increasing source of radiation exposure[J].The New England Journal of Medicine, 2007, 357(22):2277-2284.
[2] LU Z X, XIA W J, HUANG Y Q, et al.MANAS:multi-scale and multi-level neural architecture search for low-dose CT denoising[EB/OL].[2021-12-05].https://arxiv.org/pdf/2103.12995v1.pdf.
[3] BUADES A, COLL B, MOREL J M.A non-local algorithm for image denoising[C]//Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2005:60-65.
[4] KANG D, SLOMKA P, NAKAZATO R, et al.Image denoising of low-radiation dose coronary CT angiography by an adaptive block-matching 3D algorithm[EB/OL].[2021-12-05].https://www.spie.org/Publications/Proceedings/Paper/10.1117/12.2006907?SSO=1.
[5] FUMENE FERUGLIO P, VINEGONI C, GROS J, et al.Block matching 3D random noise filtering for absorption optical projection tomography[J].Physics in Medicine and Biology, 2010, 55(18):5401-5415.
[6] 李传朋, 秦品乐, 张晋京.基于深度卷积神经网络的图像去噪研究[J].计算机工程, 2017, 43(3):253-260. LI C P, QIN P L, ZHANG J J.Research on image denoising based on deep convolutional neural network[J].Computer Engineering, 2017, 43(3):253-260.(in Chinese)
[7] DONG C, LOY C C, HE K M, et al.Image super-resolution using deep convolutional networks[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2016, 38(2):295-307.
[8] 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.
[9] 章云港, 杨剑锋, 易本顺.低剂量CT图像去噪的改进型残差编解码网络[J].上海交通大学学报, 2019, 53(8):983-989. ZHANG Y G, YANG J F, YI B S.Improved residual encoder-decoder network for low-dose CT image denoising[J].Journal of Shanghai Jiao Tong University, 2019, 53(8):983-989.(in Chinese)
[10] 吕晓琪, 吴凉, 谷宇, 等.基于深度卷积神经网络的低剂量CT肺部去噪[J].电子与信息学报, 2018, 40(6):1353-1359. LÜ X Q, WU L, GU Y, et al.Low dose CT lung denoising model based on deep convolution neural network[J].Journal of Electronics & Information Technology, 2018, 40(6):1353-1359.(in Chinese)
[11] YANG Q S, YAN P K, ZHANG Y B, 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.
[12] GHOLIZADEH-ANSARI M, ALIREZAIE J, BABYN P.Deep learning for low-dose CT denoising using perceptual loss and edge detection layer[J].Journal of Digital Imaging, 2020, 33(2):504-515.
[13] LI M, HSU W, XIE X D, et al.SACNN:self-attention convolutional neural network for low-dose CT denoising with self-supervised perceptual loss network[J].IEEE Transactions on Medical Imaging, 2020, 39(7):2289-2301.
[14] GOODFELLOW I, POUGET-ABADIE J, MIRZA M, et al.Generative adversarial networks[J].Communications of the ACM, 2020, 63(11):139-144.
[15] ARJOVSKY M, CHINTALA S, BOTTOU L.Wasserstein generative adversarial networks[C]//Proceedings of International Conference on Machine Learning.Washington D.C., USA:IEEE Press, 2017:214-223.
[16] 朱斯琪, 王珏, 蔡玉芳.基于改进型循环一致性生成对抗网络的低剂量CT去噪算法[J].光学学报, 2020, 40(22):70-78. ZHU S Q, WANG J, CAI Y F.Low-dose CT denoising algorithm based on improved cycle GAN[J].Acta Optica Sinica, 2020, 40(22):70-78.(in Chinese)
[17] YUE Z, ZHAO Q, ZHANG L, et al.Dual adversarial network:toward real-world noise removal and noise generation[C]//Proceedings of European Conference on Computer Vision.Berlin, Germany:Springer, 2020:41-58.
[18] ZHANG K, ZUO W M, CHEN Y J, et al.Beyond a Gaussian denoiser:residual learning of deep CNN for image denoising[J].IEEE Transactions on Image Processing, 2017, 26(7):3142-3155.
[19] LI X M, CHEN H, QI X J, et al.H-DenseUNet:hybrid densely connected UNet for liver and tumor segmentation from CT volumes[J].IEEE Transactions on Medical Imaging, 2018, 37(12):2663-2674.
[20] HE K M, ZHANG X Y, REN S Q, et al.Deep residual learning for image recognition[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2016:770-778.
[21] LI C, XU K, LIU J, et al.Triple generative adversarial networks[EB/OL].[2021-12-05].https://arxiv.org/pdf/1912.09784v1.pdf.
[22] MA Y J, WEI B, FENG P, et al.Low-dose CT image denoising using a generative adversarial network with a hybrid loss function for noise learning[J].IEEE Access, 2020, 8:67519-67529.
[23] JOHNSON J, ALAHI A, FEI-FEI L.Perceptual losses for real-time style transfer and super-resolution[C]//Proceedings of European Conference on Computer Vision.Berlin, Germany:Springer, 2016:694-711.
[24] HE K M, CHEN X L, XIE S N, et al.Masked autoencoders are scalable vision learners[EB/OL].[2021-12-05].https://arxiv.org/abs/2111.06377.
[25] ISOLA P, ZHU J Y, ZHOU T H, et al.Image-to-image translation with conditional adversarial networks[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2017:5967-5976.
[26] ZHU J Y, PARK T, ISOLA P, et al.Unpaired image-to-image translation using cycle-consistent adversarial networks[C]//Proceedings of IEEE International Conference on Computer Vision.Washington D.C., USA:IEEE Press, 2017:2242-2251.
[27] AAPM.Low dose CT grand challenge[EB/OL].[2021-12-05].http://www.aapm.org/GrandChallenge/LowDoseCT/.
[28] LI Z, SHI W, XING Q, et al.Low-dose CT image denoising with improving WGAN and hybrid loss function[J].Computational and Mathematical Methods in Medicine, 2021, 15:2973108.
[29] WINKLER S, MOHANDAS P.The evolution of video quality measurement:from PSNR to hybrid metrics[J].IEEE Transactions on Broadcasting, 2008, 54(3):660-668.
[30] 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.
[31] XUE W F, ZHANG L, MOU X Q, et al.Gradient magnitude similarity deviation:a highly efficient perceptual image quality index[J].IEEE Transactions on Image Processing, 2014, 23(2):684-695.
[32] ZIAEI NAFCHI H, SHAHKOLAEI A, HEDJAM R, et al.Mean deviation similarity index:efficient and reliable full-reference image quality evaluator[J].IEEE Access, 2016, 4:5579-5590.

选择文件类型/文献管理软件名称

选择包含的内容