ERA-WGAT: Edge-enhanced Residual Autoencoder with a Window-based Graph Attention Convolutional Network for Low-dose CT Denoising

Overview

Journal Biomed Opt Express

Specialty Radiology

Date 2023 Feb 3

PMID 36733738

Authors

Han Liu

Peixi Liao

Hu Chen

Yi Zhang

Affiliations

Soon will be listed here.

Abstract

Computed tomography (CT) has become a powerful tool for medical diagnosis. However, minimizing X-ray radiation risk for the patient poses significant challenges to obtain suitable low dose CT images. Although various low-dose CT methods using deep learning techniques have produced impressive results, convolutional neural network based methods focus more on local information and hence are very limited for non-local information extraction. This paper proposes ERA-WGAT, a residual autoencoder incorporating an edge enhancement module that performs convolution with eight types of learnable operators providing rich edge information and a window-based graph attention convolutional network that combines static and dynamic attention modules to explore non-local self-similarity. We use the compound loss function that combines MSE loss and multi-scale perceptual loss to mitigate the over-smoothing problem. Compared with current low-dose CT denoising methods, ERA-WGAT confirmed superior noise suppression and perceived image quality.

Citing Articles

A systematic review of deep learning-based denoising for low-dose computed tomography from a perceptual quality perspective.

Kim W, Jeon S, Byun G, Yoo H, Choi J Biomed Eng Lett. 2024; 14(6):1153-1173.

PMID: 39465112 PMC: 11502640. DOI: 10.1007/s13534-024-00419-7.

References

Berrington de Gonzalez A, Mahesh M, Kim K, Bhargavan M, Lewis R, Mettler F . Projected cancer risks from computed tomographic scans performed in the United States in 2007. Arch Intern Med. 2009; 169(22):2071-7. PMC: 6276814. DOI: 10.1001/archinternmed.2009.440. View

Wang J, Li T, Lu H, Liang Z . Penalized weighted least-squares approach to sinogram noise reduction and image reconstruction for low-dose X-ray computed tomography. IEEE Trans Med Imaging. 2006; 25(10):1272-83. PMC: 1619874. DOI: 10.1109/42.896783. View

Balda M, Hornegger J, Heismann B . Ray contribution masks for structure adaptive sinogram filtering. IEEE Trans Med Imaging. 2012; 31(6):1228-39. DOI: 10.1109/TMI.2012.2187213. View

Fumene Feruglio P, Vinegoni C, Gros J, Sbarbati A, Weissleder R . Block matching 3D random noise filtering for absorption optical projection tomography. Phys Med Biol. 2010; 55(18):5401-15. PMC: 2934766. DOI: 10.1088/0031-9155/55/18/009. View

Yang Q, Yan P, Zhang Y, Yu H, Shi Y, Mou X . Low-Dose CT Image Denoising Using a Generative Adversarial Network With Wasserstein Distance and Perceptual Loss. IEEE Trans Med Imaging. 2018; 37(6):1348-1357. PMC: 6021013. DOI: 10.1109/TMI.2018.2827462. View

Chen H, Zhang Y, Kalra M, Lin F, Chen Y, Liao P . Low-Dose CT With a Residual Encoder-Decoder Convolutional Neural Network. IEEE Trans Med Imaging. 2017; 36(12):2524-2535. PMC: 5727581. DOI: 10.1109/TMI.2017.2715284. View

Chen H, Zhang Y, Zhang W, Liao P, Li K, Zhou J . Low-dose CT via convolutional neural network. Biomed Opt Express. 2017; 8(2):679-694. PMC: 5330597. DOI: 10.1364/BOE.8.000679. View

Fan F, Shan H, Kalra M, Singh R, Qian G, Getzin M . Quadratic Autoencoder (Q-AE) for Low-Dose CT Denoising. IEEE Trans Med Imaging. 2020; 39(6):2035-2050. PMC: 7376975. DOI: 10.1109/TMI.2019.2963248. View

Sheng K, Gou S, Wu J, Qi S . Denoised and texture enhanced MVCT to improve soft tissue conspicuity. Med Phys. 2014; 41(10):101916. DOI: 10.1118/1.4894714. View

10.

Zhang Y, Xi Y, Yang Q, Cong W, Zhou J, Wang G . Spectral CT Reconstruction with Image Sparsity and Spectral Mean. IEEE Trans Comput Imaging. 2017; 2(4):510-523. PMC: 5637560. DOI: 10.1109/TCI.2016.2609414. View

11.

Zhang Y, Wang Y, Zhang W, Lin F, Pu Y, Zhou J . Statistical iterative reconstruction using adaptive fractional order regularization. Biomed Opt Express. 2016; 7(3):1015-29. PMC: 4866445. DOI: 10.1364/BOE.7.001015. View

12.

Ma J, Huang J, Feng Q, Zhang H, Lu H, Liang Z . Low-dose computed tomography image restoration using previous normal-dose scan. Med Phys. 2011; 38(10):5713-31. PMC: 3298559. DOI: 10.1118/1.3638125. View

13.

Chen Y, Gao D, Nie C, Luo L, Chen W, Yin X . Bayesian statistical reconstruction for low-dose X-ray computed tomography using an adaptive-weighting nonlocal prior. Comput Med Imaging Graph. 2009; 33(7):495-500. DOI: 10.1016/j.compmedimag.2008.12.007. View

14.

Li Z, Yu L, Trzasko J, Lake D, Blezek D, Fletcher J . Adaptive nonlocal means filtering based on local noise level for CT denoising. Med Phys. 2014; 41(1):011908. DOI: 10.1118/1.4851635. View

15.

Manduca A, Yu L, Trzasko J, Khaylova N, Kofler J, McCollough C . Projection space denoising with bilateral filtering and CT noise modeling for dose reduction in CT. Med Phys. 2009; 36(11):4911-9. PMC: 4108640. DOI: 10.1118/1.3232004. View

16.

Wolterink J, Leiner T, Viergever M, Isgum I . Generative Adversarial Networks for Noise Reduction in Low-Dose CT. IEEE Trans Med Imaging. 2017; 36(12):2536-2545. DOI: 10.1109/TMI.2017.2708987. View

17.

Sidky E, Pan X . Image reconstruction in circular cone-beam computed tomography by constrained, total-variation minimization. Phys Med Biol. 2008; 53(17):4777-807. PMC: 2630711. DOI: 10.1088/0031-9155/53/17/021. View

18.

Valsesia D, Fracastoro G, Magli E . Deep Graph-Convolutional Image Denoising. IEEE Trans Image Process. 2020; PP. DOI: 10.1109/TIP.2020.3013166. View

19.

Chen Y, Yin X, Shi L, Shu H, Luo L, Coatrieux J . Improving abdomen tumor low-dose CT images using a fast dictionary learning based processing. Phys Med Biol. 2013; 58(16):5803-20. DOI: 10.1088/0031-9155/58/16/5803. View

20.

McCollough C, Bartley A, Carter R, Chen B, Drees T, Edwards P . Low-dose CT for the detection and classification of metastatic liver lesions: Results of the 2016 Low Dose CT Grand Challenge. Med Phys. 2017; 44(10):e339-e352. PMC: 5656004. DOI: 10.1002/mp.12345. View