Sparse reconstruction of CT images based on Uformer with fused channel attention

doi:10.11772/j.issn.1001-9081.2022081242

Abstract

Abstract:

Concerning the problem of streak artifacts generated in the sparse reconstruction of analytic method， a Channel Attention U-shaped Transformer （CA-Uformer） was proposed to achieve high-precision Computed Tomography （CT） sparse reconstruction. In CA-Uformer， channel attention and spatial attention in Transformer were fused， and with the dual-attention mechanism， image detail information was easier learnt by the network； an excellent U-shaped architecture was adopted to fuse multi-scale image information； a forward feedback network design was implemented by using convolutional operations， which further coupled the local information association ability of Convolutional Neural Network （CNN） and the global information capturing ability of Transformer. Experimental results show that CA-Uformer has the Peak Signal-to-Noise Ratio （PSNR）， Structural Similarity （SSIM） 3.27 dB and 3.14% higher， and Root Mean Square Error （RMSE） 35.29% lower than the classical U-Net， which is a significant improvement. It can be seen that CA-Uformer has sparse reconstruction with higher precision and better ability to suppress artifacts.

Key words: Computed Tomography (CT), sparse reconstruction, strip artifact, Transformer, channel attention

摘要：

针对解析法稀疏重建中产生的条状伪影问题，提出一种融合通道注意力的U型Transformer（CA-Uformer），以实现高精度计算机断层成像（CT）的稀疏重建。CA-Uformer融合了通道注意力和Transformer中的空间注意力，双注意力机制使网络更容易学习到图像细节信息；采用优秀的U型架构融合多尺度图像信息；采用卷积操作实现前向反馈网络设计，从而进一步耦合卷积神经网络（CNN）的局部信息关联能力和Transformer的全局信息捕捉能力。实验结果表明，与经典U-Net相比，CA-Uformer的峰值信噪比（PSNR）、结构相似性（SSIM）提高了3.27 dB、3.14%，均方根误差（RMSE）降低了35.29%，提升效果明显。可见，CA-Uformer稀疏重建精度更高，压制伪影能力更强。

关键词: 计算机断层成像, 稀疏重建, 条状伪影, Transformer, 通道注意力

CLC Number:

TP391.41

Mengmeng CHEN, Zhiwei QIAO. Sparse reconstruction of CT images based on Uformer with fused channel attention[J]. Journal of Computer Applications, 2023, 43(9): 2948-2954.

陈蒙蒙, 乔志伟. 基于融合通道注意力的Uformer的CT图像稀疏重建[J]. 《计算机应用》唯一官方网站, 2023, 43(9): 2948-2954.

Figures/Tables 15

References 25

1	SIDKY E Y， KAO C M， PAN X C. Accurate image reconstruction from few-views and limited-angle data in divergent-beam CT［J］. Journal of X-Ray Science and Technology， 2006， 14（2）： 119-139.
2	乔志伟. 总变差约束的数据分离最小图像重建模型及其 Chambolle-Pock求解算法［J］. 物理学报， 2018， 67（19）：338-351. 10.7498/aps.67.20180839
	QIAO Z W. The total variation constrained data divergence minimization model for image reconstruction and its Chambolle-Pock solving algorithm［J］. Acta Physica Sinica，2018， 67（19）：338-351. 10.7498/aps.67.20180839
3	BROOKS T， MILDENHALL B， XUE T F， et al. Unprocessing images for learned raw denoising［C］// Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2019： 11028-11037. 10.1109/cvpr.2019.01129
4	LEE D， YOO J， YE J C. Deep residual learning for compressed sensing MRI［C］// Proceedings of the IEEE 14th International Symposium on Biomedical Imaging. Piscataway： IEEE， 2017： 15-18. 10.1109/isbi.2017.7950457
5	WOO S， PARK J， LEE J Y， et al. CBAM： convolutional block attention module［C］// Proceedings of the 2018 European Conference on Computer Vision， LNCS 11211. Cham： Springer， 2018： 3-19.
6	HUANG G， LIU Z， L van der MAATEN， et al. Densely connected convolutional networks［C］// Proceedings of the 2017 IEEE conference on computer vision and pattern recognition. Piscataway： IEEE， 2017： 2261-2269. 10.1109/cvpr.2017.243
7	GOODFELLOW I J， POUGET-ABADIE J， MIRZA M， et al. Generative adversarial networks［C］// Proceedings of the 27th International Conference on Neural Information Processing Systems - Volume 2. Cambridge： MIT Press， 2014： 2672-2680.
8	RONNEBERGER O， FISCHER P， BROX T. U-net： convolutional networks for biomedical image segmentation［C］// Proceedings of the 2015 International Conference on Medical Image Computing and Computer-Assisted Intervention， LNCS 9351. Cham： Springer， 2015： 234-241.
9	JIN K H， McCANN M T， FROUSTEY E， et al. Deep convolutional neural network for inverse problems in imaging［J］. IEEE Transactions on Image Processing， 2017， 26（9）： 4509-4522. 10.1109/tip.2017.2713099
10	PAN X C， SIDKY E Y， VANNIER M. Why do commercial CT scanners still employ traditional， filtered back-projection for image reconstruction？［J］. Inverse Problems， 2009， 25（12）： No.123009. 10.1088/0266-5611/25/12/123009
11	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. 10.1109/tmi.2017.2715284
12	HE K M， ZHANG X Y， REN S Q， et al. Deep residual learning for image recognition［C］// Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2016： 770-778. 10.1109/cvpr.2016.90
13	HU J， SHEN L， SUN G. Squeeze-and-excitation networks［C］// Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2018： 7132-7141. 10.1109/cvpr.2018.00745
14	ZHANG Z C， LIANG X K， DONG X， et al. A sparse-view CT reconstruction method based on combination of DenseNet and deconvolution［J］. IEEE Transactions on Medical Imaging， 2018， 37（6）： 1407-1417. 10.1109/tmi.2018.2823338
15	XU L， REN J S J， LIU C， et al. Deep convolutional neural network for image deconvolution［C］// Proceedings of the 27th International Conference on Neural Information Processing Systems - Volume 1. Cambridge： MIT Press， 2014： 1790-1798.
16	WOLTERINK J M， 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. 10.1109/tmi.2017.2708987
17	VASWANI A， SHAZEER N， PARMAR N， et al. Attention is all you need［C］// Proceedings of the 31st International Conference on Neural Information Processing Systems. Red Hook， NY： Curran Associates Inc.， 2017：6000-6010.
18	DOSOVITSKIY A， BEYER L， KOLESNIKOV A， et al. An image is worth 16x16 words： Transformers for image recognition at scale［EB/OL］. （2021-06-03）［2022-10-22］..
19	HUANG Z L， WANG X G， HUANG L C， et al. CCNet： criss-cross attention for semantic segmentation［C］// Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision. Piscataway： IEEE， 2019： 603-612. 10.1109/iccv.2019.00069
20	LIU Z， LIN Y， CAO Y， et al. Swin Transformer： hierarchical vision transformer using shifted windows［C］// Proceedings of the 2021 IEEE/CVF International Conference on Computer Vision. Piscataway： IEEE， 2021： 9992-10002. 10.1109/iccv48922.2021.00986
21	WANG X L， GIRSHICK R， GUPTA A， et al. Non-local neural networks［C］// Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2018： 7794-7803. 10.1109/cvpr.2018.00813
22	WANG Z D， CUN X D， BAO J M， et al. Uformer： a general U-shaped Transformer for image restoration［C］// Proceedings of the 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2022： 17662-17672. 10.1109/cvpr52688.2022.01716
23	PENG Z L， HUANG W， GU S Z， et al. Conformer： local features coupling global representations for visual recognition［C］// Proceedings of the 2021 IEEE/CVF International Conference on Computer Vision. Piscataway： IEEE， 2021： 357-366. 10.1109/iccv48922.2021.00042
24	YUAN K， GUO S P， LIU Z W， et al. Incorporating convolution designs into visual Transformers［C］// Proceedings of the 2021 IEEE/CVF International Conference on Computer Vision. Piscataway： IEEE， 2021： 559-568. 10.1109/iccv48922.2021.00062
25	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. 10.1109/tip.2017.2662206

算法	PSNR/dB	SSIM	RMSE	参数量/10⁶	浮点运算量/GFLOPs	重建时间/s
DnCNN	32.18	0.939	0.026	0.14	18.45	0.18
RED-CNN	32.96	0.963	0.023	0.21	24.60	0.18
U-Net	36.03	0.955	0.017	7.77	27.42	0.18
FBPConvNet	37.69	0.962	0.014	9.16	29.60	0.19
Uformer	38.54	0.982	0.012	20.77	82.01	0.31
CA-Uformer	39.30	0.985	0.011	76.61	311.25	0.30

算法	PSNR/dB	SSIM	RMSE	参数量/10⁶	浮点运算量/GFLOPs	重建时间/s
DnCNN	32.18	0.939	0.026	0.14	18.45	0.18
RED-CNN	32.96	0.963	0.023	0.21	24.60	0.18
U-Net	36.03	0.955	0.017	7.77	27.42	0.18
FBPConvNet	37.69	0.962	0.014	9.16	29.60	0.19
Uformer	38.54	0.982	0.012	20.77	82.01	0.31
CA-Uformer	39.30	0.985	0.011	76.61	311.25	0.30

稀疏角度数	PSNR/dB	SSIM	RMSE
15	33.11	0.956	0.022
30	36.40	0.974	0.015
60	39.30	0.985	0.011
90	40.36	0.988	0.010

稀疏角度数	PSNR/dB	SSIM	RMSE
15	33.11	0.956	0.022
30	36.40	0.974	0.015
60	39.30	0.985	0.011
90	40.36	0.988	0.010

模块位置设计	PSNR/dB	SSIM	RMSE
CA-Identity	38.04	0.984	0.013
Transformer-CA	38.57	0.983	0.012
CA-Uformer	39.30	0.985	0.011