Limited-angle Computed Tomography imaging suffers from severe artifacts in the reconstructed image due to incomplete projection data. Deep learning methods have been developed currently to address the challenges of robustness and low contrast of the limited-angle CT reconstruction with a relatively effective way.
Objective
To improve the low contrast of the current limited-angle CT reconstruction image, enhance the robustness of the reconstruction method and the contrast of the limited-angle image.
Method
In this paper, we proposed a limited-angle CT reconstruction method that combining the Fourier domain reweighting and wavelet domain enhancement, which fused information from different domains, thereby getting high-resolution reconstruction images.
Results
We verified the feasibility and effectiveness of the proposed solution through experiments, and the reconstruction results are improved compared with the state-of-the-art methods.
Conclusions
The proposed method enhances some features of the original image domain data from different domains, which is beneficial to the reasonable diffusion and restoration of diffuse detail texture features.
NiklasonLChristianBNiklasonL, et al.Digital tomosynthesis in breast imaging. Radiology1997; 205: 399–406.
2.
TingbergA. X-ray tomosynthesis: a review of its use for breast and chest imaging. Radiat Prot Dosimetry2010; 139: 100–107.
3.
BacharGSiewerdsenJDalyM, et al.Image quality and localization accuracy in c-arm tomosynthesis-guided head and neck surgery. Med Phys2007; 34: 4664–4677.
4.
FahrigRDixonRPayneT, et al.Dose and image quality for a cone-beam c-arm CT system. Med Phys2006; 33: 4541–4550.
5.
ChoMKimHYounH, et al.A feasibility study of digital tomosynthesis for volumetric dental imaging. J Instrum2012; 7: P03007.
6.
GuoYZengLWangC, et al.Image reconstruction model for the exterior problem of computed tomography based on weighted directional total variation. Appl Math Model2017; 52: 358–377.
7.
SidkyEYKaoCMPanX. Accurate image reconstruction from few-views and limited-angle data in divergent-beam CT. J Xray Sci Technol2006; 14: 119–139.
8.
YuHWangG. Compressed sensing based interior tomography. Phys Med Biol2009; 54: 2791–2805.
9.
WeiYZengL. gradient minimization based image reconstruction for limited-angle computed tomography. PLoS ONE2015; 10: 1–15.
10.
XiaodongZPalenstijnWJBatenburgKJ. Tvr-dart: a more robust algorithm for discrete tomography from limited projection data with automated gray value estimation. IEEE Trans Image Process2016; 25: 455–468.
11.
XuJZhaoYLiH, et al.An image reconstruction model regularized by edge-preserving diffusion and smoothing for limited-angle computed tomography. Inverse Probl2019; 35: 085004.
12.
XuMHuDLuoF, et al.Limited-angle x-ray CT reconstruction using image gradient -norm with dictionary learning. IEEE Trans Rad Plasma Med Sci2020; 5: 78–87.
13.
HuangYTaubmannOHuangX, et al.Scale-space anisotropic total variation for limited angle tomography. IEEE Trans Rad Plasma Med Sci2018; 2: 307–314.
14.
VenkatakrishnanSVDrummyLFJacksonMA, et al.A model based iterative reconstruction algorithm for high angle annular dark field-scanning transmission electron microscope (haadf-stem) tomography. IEEE Trans Image Process2013; 22: 4532–4544.
15.
ChenHZhangYKalraMK, et al.Low-dose ct with a residual encoder-decoder convolutional neural network. IEEE Trans Med Imaging2017; 36: 2524–2535.
16.
JinKHMcCannMTFrousteyE, et al.Deep convolutional neural network for inverse problems in imaging. IEEE Trans Image Process2017; 26: 4509–4522.
17.
KangEMinJYeJC. A deep convolutional neural network using directional wavelets for low-dose x-ray ct reconstruction. Med Phys2017; 44: e360–e375.
18.
LiangTJinYLiY, et al.Edcnn: Edge enhancement-based densely connected network with compound loss for low-dose ct denoising. In: 2020 15th IEEE International conference on signal processing (ICSP), Volume 1, 2020, pp.193–198. IEEE.
19.
ZhangYLvTGeRet al.CD-net: comprehensive domain network with spectral complementary for dect sparse-view reconstruction. IEEE Trans Comput Imaging2021; 7: 436–447.
20.
AnirudhRKimHThiagarajanJJ, et al.Lose the views: limited angle ct reconstruction via implicit sinogram completion. In: Proceedings of the IEEE conference on computer vision and pattern recognition, 2018, pp.6343–6352.
21.
GuptaHJinKHNguyenHQ, et al.CNN-based projected gradient descent for consistent CT image reconstruction. IEEE Trans Med Imaging2018; 37: 1440–1453.
22.
HuangYHuangXTaubmannO, et al.Restoration of missing data in limited angle tomography based on helgason–ludwig consistency conditions. Biomed Phys Eng Exp2017; 3: 035015.
23.
AntunVRennaFPoonC, et al.On instabilities of deep learning in image reconstruction and the potential costs of ai. Proc Natl Acad Sci2020; 117: 30088–30095.
24.
CubukEDZophBSchoenholzSS, et al.Intriguing properties of adversarial examples. arXiv preprint arXiv:1711.02846, 2017.
25.
HuangYWürflTBreiningerK, et al.Some investigations on robustness of deep learning in limited angle tomography. In: Bildverarbeitung für die Medizin 2019: Algorithmen–systeme–anwendungen. Proceedings des workshops vom 17. bis 19. März 2019 in Lübeck, 2019, pp.21–21. Springer.
26.
ChenHZhangYChenY, et al.Learn: learned experts’ assessment-based reconstruction network for sparse-data ct. IEEE Trans Med Imaging2018; 37: 1333–1347.
27.
WangJZengLWangC, et al.Admm-based deep reconstruction for limited-angle CT. Phys Med Biol2019; 64: 115011.
28.
ZhangHDongBLiuB. Jsr-net: a deep network for joint spatial-radon domain CT reconstruction from incomplete data. In: ICASSP 2019-2019 IEEE International conference on acoustics, speech and signal processing (ICASSP), 2019, pp.3657–3661. IEEE.
29.
RomanoYEladMMilanfarP. The little engine that could: regularization by denoising (red). SIAM J Imaging Sci2017; 10: 1804–1844.
30.
VenkatakrishnanSVBoumanCAWohlbergB. Plug-and-play priors for model based reconstruction. In: 2013 IEEE global conference on signal and information processing, 2013, pp.945–948. IEEE.
31.
SreehariSVenkatakrishnanSVWohlbergB, et al.Plug-and-play priors for bright field electron tomography and sparse interpolation. IEEE Trans Comput Imaging2016; 2: 408–423.
32.
ZhangKLiYZuoW, et al.Plug-and-play image restoration with deep denoiser prior. IEEE Trans Pattern Anal Mach Intell2021; 44: 6360–6376.
33.
Usman GhaniMClem KarlW. Integrating data and image domain deep learning for limited angle tomography using consensus equilibrium. In: Proceedings of the IEEE/CVF international conference on computer vision workshops, 2019, pp.0–0.
34.
WürflTHoffmannMChristleinV, et al.Deep learning computed tomography: learning projection-domain weights from image domain in limited angle problems. IEEE Trans Med Imaging2018; 37: 1454–1463.
35.
ZhouBChenXZhouSK, et al.Dudodr-net: dual-domain data consistent recurrent network for simultaneous sparse view and metal artifact reduction in computed tomography. Med Image Anal2022; 75: 102289.
36.
LiuJAnirudhRThiagarajanJJ, et al.Dolce: a model-based probabilistic diffusion framework for limited-angle CT reconstruction. In: Proceedings of the IEEE/CVF international conference on computer vision, 2023, pp.10498–10508.
37.
XieJShaoHCLiY, et al.Prior frequency guided diffusion model for limited angle (la)-cbct reconstruction. Phys Med Biol2024; 69: 135008.
38.
ZhangJMaoHWangX, et al.Wavelet-inspired multi-channel score-based model for limited-angle CT reconstruction. IEEE Trans Med Imaging2024; 43: 3436–3448.
39.
DhariwalPNicholA. Diffusion models beat gans on image synthesis. Adv Neural Inf Process Syst2021; 34: 8780–8794.
40.
WangJLyuZLinD, et al.Guided diffusion model for adversarial purification. arXiv preprint arXiv:2205.14969, 2022.
41.
AdlerJÖktemO. Learned primal-dual reconstruction. IEEE Trans Med Imaging2018; 37: 1322–1332.
42.
GaoQDingRWangL, et al.Lrip-net: low-resolution image prior-based network for limited-angle CT reconstruction. IEEE Trans Rad Plasma Med Sci2022; 7: 163–174.
43.
HuDZhangYLiuJ, et al.Dior: deep iterative optimization-based residual-learning for limited-angle CT reconstruction. IEEE Trans Med Imaging2022; 41: 1778–1790.
44.
LiuJSunYGanW, et al.Sgd-net: efficient model-based deep learning with theoretical guarantees. IEEE Trans Comput Imaging2021; 7: 598–610.
45.
ZhouBZhouSKDuncanJS, et al.Limited view tomographic reconstruction using a cascaded residual dense spatial-channel attention network with projection data fidelity layer. IEEE Trans Med Imaging2021; 40: 1792–1804.
NicholAQDhariwalP. Improved denoising diffusion probabilistic models. In: International conference on machine learning, 2021, pp.8162–8171. PMLR.
48.
SiCHuangZJiangY, et al.Freeu: Free lunch in diffusion u-net. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2024, pp.4733–4743.
49.
WangXDarrellTRambhatlaSS, et al.Instancediffusion: Instance-level control for image generation. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2024, pp.6232–6242.
50.
PhungHDaoQTranA. Wavelet diffusion models are fast and scalable image generators. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2023, pp.10199–10208.
51.
HestenesMRStiefelE, et al.Methods of conjugate gradients for solving linear systems. NBS Washington, DC, 1952.
52.
McColloughCHBartleyACCarterREet al.Low-dose CT for the detection and classification of metastatic liver lesions: results of the 2016 low dose ct grand challenge. Med Phys2017; 44: e339–e352.
53.
KimHChampleyK. Differentiable forward projector for x-ray computed tomography. In: ICML 2023 Workshop on differentiable almost everything: differentiable relaxations, algorithms, operators, and simulators, 2023, https://openreview.net/forum?id=dX8khDFGHv.
54.
OktayO. Attention u-net: learning where to look for the pancreas. arXiv preprint arXiv:1804.03999, 2018.
55.
SchlemperJOktayOSchaapM, et al.Attention gated networks: Learning to leverage salient regions in medical images. Med Image Anal2019; 53: 197–207.
