Although computed tomography (CT) is widely employed in disease detection, X-ray radiation may pose a risk to the health of patients. Reducing the projection views is a common method, however, the reconstructed images often suffer from streak artifacts.
Purpose
In previous related works, it can be found that the convolutional neural network (CNN) is proficient in extracting local features, while the Transformer is adept at capturing global information. To suppress streak artifacts for sparse-view CT, this study aims to develop a method that combines the advantages of CNN and Transformer.
Methods
In this paper, we propose a Multi-Attention and Dual-Branch Feature Aggregation U-shaped Transformer network (MAFA-Uformer), which consists of two branches: CNN and Transformer. Firstly, with a coordinate attention mechanism, the Transformer branch can capture the overall structure and orientation information to provide a global context understanding of the image under reconstruction. Secondly, the CNN branch focuses on extracting crucial local features of images through channel spatial attention, thus enhancing detail recognition capabilities. Finally, through a feature fusion module, the global information from the Transformer and the local features from the CNN are integrated effectively.
Results
Experimental results demonstrate that our method achieves outstanding performance in terms of peak signal-to-noise ratio (PSNR), structural similarity (SSIM), and root mean square error (RMSE). Compared with Restormer, our model achieves significant improvements: PSNR increases by 0.76 dB, SSIM improves by 0.44%, and RMSE decreases by 8.55%.
Conclusion
Our method not only effectively suppresses artifacts but also better preserves details and features, thereby providing robust support for accurate diagnosis of CT images.
WangGYeJCDe ManB. Deep learning for tomographic image reconstruction. Nat Mach Intell2020; 2: 737–748.
2.
LiZLiuYZhangP, et al.Decomposition iteration strategy for low-dose CT denoising. J Xray Sci Technol2024; 32: 493–512.
3.
de GonzalezABDarbyS. Risk of cancer from diagnostic X-rays: estimates for the UK and 14 other countries. Lancet2004; 363: 345–351.
4.
FengZCaiAWangY, et al.Dual residual convolutional neural network (DRCNN) for low-dose CT imaging. J Xray Sci Technol2021; 29: 91–109.
5.
GuYLiuYLiuW, et al.Sparse angle CT reconstruction based on group sparse representation. J Xray Sci Technol2022; 30: 1085–1097.
6.
HuangYWanQChenZ, et al.An iterative reconstruction method for sparse-projection data for low-dose CT. J Xray Sci Technol2021; 29: 797–812.
7.
PanXSidkyEYVannierM. Why do commercial CT scanners still employ traditional, filtered back-projection for image reconstruction?Inverse Prob2009; 25. doi: https://doi.org/10.1088/0266-5611/25/12/123009.
8.
NiuSGaoYBianZ, et al.Sparse-view x-ray CT reconstruction via total generalized variation regularization. Phys Med Biol2014; 59: 2997–3017. doi: https://doi.org/10.1088/0031-9155/59/12/2997.
9.
ZhaoXJiangCZhangQ, et al.Iterative image reconstruction for sparse-view CT via total variation regularization and dictionary learning. J Xray Sci Technol2019; 27: 573–590.
10.
LeeHLeeJKimH, et al.Deep-neural-network-based sinogram synthesis for sparse-view CT image reconstruction. IEEE Transactions on Radiation and Plasma Medical Sciences2019; 3: 109–119.
11.
GuanBYangCZhangL, et al.Generative modeling in sinogram domain for sparse-view CT reconstruction. IEEE Transactions on Radiation and Plasma Medical Sciences2024; 8: 195–207.
12.
HuangZZhangJZhangY, et al.DU-GAN: generative adversarial networks with dual-domain U-Net-based discriminators for Low-Dose CT Denoising. IEEE Trans Instrum Meas2022; 71. doi: https://doi.org/10.1109/TIM.2021.3128703.
ChenHZhangYChenY, et al.LEARN: learned experts’ assessment-based reconstruction network for sparse-data CT. IEEE Trans Med Imaging2018; 37: 1333–1347.
15.
YangYSunJLiH, et al.Deep ADMM-net for compressive sensing MRI. In: Advances in Neural Information Processing Systems, 2016, pp.10–18.
16.
ChenHZhangYZhangW, et al.Low-dose CT denoising with convolutional neural network. 2017 IEEE 14th International Symposium on Biomedical Imaging2017: 143–146. doi: https://doi.org/10.1109/ISBI.2017.7950488.
17.
LiuYYanRLiuY, et al.Enhancement based convolutional dictionary network with adaptive window for low-dose CT denoising. J Xray Sci Technol2023; 31: 1165–1187.
18.
JinKHMcCannMTFrousteyE, et al.Deep convolutional neural network for inverse problems in imaging. IEEE Trans Image Process2017; 26: 4509–4522.
19.
ChenHZhangYKalraMK, et al.Low-dose CT with a residual encoder-decoder convolutional neural network. IEEE Trans Med Imaging2017; 36: 2524–2535.
20.
XuLRenJSLiuC, et al.Deep convolutional neural network for image deconvolution. Adv Neural Inf Process Syst2014; 2: 1790–1798.
21.
ZhangZLiangXDongX, et al.A sparse-view CT reconstruction method based on combination of DenseNet and deconvolution. IEEE Trans Med Imaging2018; 37: 1407–1417.
22.
HuangGLiuZVan Der MaatenL, et al.Densely connected convolutional networks. In: Proceedings of the IEEE conference on computer vision and pattern recognition, 2017, pp.2261–2269. doi: https://doi.org/10.1109/CVPR.2017.243.
23.
YangQYanPZhangY, et al.Low-dose CT image denoising using a generative adversarial network with wasserstein distance and perceptual loss. IEEE Trans Med Imaging2018; 37: 1348–1357.
24.
YanRLiuYLiuY, et al.Image denoising for low-dose CT via convolutional dictionary learning and neural network. IEEE Trans Comput Imaging2023; 9: 83–93.
25.
VaswaniAShazeerNParmarN, et al.Attention is all you need. Adv Neural Inf Process Syst2017: 5999–6009.
26.
LinXSunSHuangW, et al.EAPT: efficient attention pyramid transformer for image processing. IEEE Trans Multimed2023; 25: 50–61.
27.
YinHMaS. CSformer: cross-scale features fusion based transformer for image denoising. IEEE Signal Process Lett2022; 29: 1809–1813.
28.
DosovitskiyABeyerLKolesnikovA, et al.An image is worth 16 ( 16 words: Transformers for image recognition at scale. In: 9th International Conference on Learning Representations2021.
29.
LiuZLinYCaoY, et al.Swin transformer: hierarchical vision transformer using shifted windows. In: Proceedings of the IEEE International Conference on Computer Vision, 2021, pp.9992–10002. doi: https://doi.org/10.1109/ICCV48922.2021.00986.
30.
WangZCunXBaoJ, et al.Uformer: a general u-shaped transformer for image restoration. In: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognitio, 2022, pp.17662–17672. doi: https://doi.org/10.1109/CVPR52688.2022.01716.
31.
ZamirSWAroraAKhanS, et al.Restormer: efficient transformer for high-resolution image restoration. In: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2022, pp.5718–5729. doi: https://doi.org/10.1109/CVPR52688.2022.00564.
32.
ZhangZYuLLiangX, et al.TransCT: dual-path transformer for low dose computed tomography. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)2021; 12906: 55–64.
33.
YuanJZhouFGuoZ, et al.HCformer: hybrid CNN-transformer for LDCT image denoising. J Digit Imaging2023; 36: 2290–2305.
34.
XueTMaP. TC-net: transformer combined with cnn for image denoising. Applied Intelligence2023; 53: 6753–6762.
HouQZhouDFengJ. Coordinate attention for efficient mobile network design. In: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2021, pp.13708–13717. doi: https://doi.org/10.1109/CVPR46437.2021.01350.
37.
WooSParkJLeeJY, et al.Cbam: convolutional block attention module. Proceedings of the European Conference on Computer Vision (ECCV)2018; 11211: 3–19.
38.
GajeraBKapilSRZiaeiD, et al.CT-scan denoising using a charbonnier loss generative adversarial network. IEEE Access2021; 9: 84093–84109.
39.
LiSLiQLiR, et al.An adaptive self-guided wavelet convolutional neural network with compound loss for low-dose CT denoising. Biomed Signal Process Control2022; 75. doi: https://doi.org/10.1016/j.bspc.2022.103543
40.
Huynh-ThuQGhanbariM. Scope of validity of PSNR in image/video quality assessment. Electron Lett2008; 44: 800–801.
41.
HoreAZiouD. Image quality metrics: PSNR vs. SSIM. In: 20th international conference on pattern recognition, 2010, pp.2366–2369. doi: https://doi.org/10.1109/ICPR.2010.579.
42.
ChaiTDraxlerRR. Root mean square error (RMSE) or mean absolute error (MAE)? – arguments against avoiding RMSE in the literature. Geoscientific Model Development2014; 7: 1247–1250.
43.
ZhangKZuoWChenY, et al.Beyond a Gaussian denoiser: residual learning of deep cnn for image denoising. IEEE Trans Image Process2017; 26: 3142–3155.
44.
RonnebergerOFischerPBroxT. U-net: convolutional networks for biomedical image segmentation. 18th International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI)2015; 9351: pp. 234–241.
