Low-dose CT image denoising method based on generative adversarial network

Abstract

In order to solve the problems of artifacts and noise in low-dose computed tomography (CT) images in clinical medical diagnosis, an improved image denoising algorithm under the architecture of generative adversarial network (GAN) was proposed. First, a noise model based on style GAN2 was constructed to estimate the real noise distribution, and the noise information similar to the real noise distribution was generated as the experimental noise data set. Then, a network model with encoder-decoder architecture as the core based on GAN idea was constructed, and the network model was trained with the generated noise data set until it reached the optimal value. Finally, the noise and artifacts in low-dose CT images could be removed by inputting low-dose CT images into the denoising network. The experimental results showed that the constructed network model based on GAN architecture improved the utilization rate of noise feature information and the stability of network training, removed image noise and artifacts, and reconstructed image with rich texture and realistic visual effect.

Keywords

low-dose CT image / generative adversarial network / noise and artifacts / encoder-decoder / atrous spatial pyramid pooling (ASPP)

References

[1] TANG Y, CUI X Y, ZHANG Q, et al. Denoising algorithm for low-dose CT projection based on partial differential equation. Computer Application & Software, 2014, 31(3): 179-182.

[2] ZHANG H Y, TENG G J, CAO B, et al. Advances in X-ray imaging technology. Scientia Sinica Vitae, 2020, 50(11): 1202-1212.

[3] LI T F, LI X, YANG Y, et al. Simultaneous reduction of radiation dose and scatter for CBCT by using collimators. Medical Physics, 2013, 40(12): 121913.

[4] CHAI N, WANG S J, CHEN L J, et al. Low-dose micro-CT imaging method based on progressive network processing. Computerized Tomography Theory and Applications, 2020, 29(4): 435-446,.

[5] YU L F, MANDUCA A, TRZASKO J D, et al. Sinogram smoothing with bilateral filtering for low-dose CT//Medical Imaging 2008: Physics of Medical Imaging, March 18, 2008, San Diego, CA. Bellingham: SPIE, 2008: 768-775.

[6] DEMIRKAYA O. Reduction of noise and image artifacts in computed tomography by nonlinear filtration of the projection images//Medical Imaging 2001: Physics of Medical Imaging, February 16, 2001, San Diego, CA, USA. Bellingham: SPIE, 2001: 917-923.

[7] KACHELRIEß M, WATZKE O, KALENDER W A. Generalized multi-dimensional adaptive filtering for conventional and spiral single-slice, multi-slice, and cone-beam CT. Medical Physics, 2001, 28(4): 475-490.

[8] WANG J, LU H B, LI T F,et al. Sinogram noise reduction for 1ow-dose CT by statistics-based onlinear filters//Medical Imaging 2005: Image Processing, February 13-15, 2005, San Diego, CA, USA. Bellingham: SPIE, 2005: 2058-2066.

[9] LIU Y, ZHANG Q, GUI Z G. Noise reduction for low-dose CT sinogram based on fuzzy entropy. Journal of Electronics & Information Technology, 2014, 35(6): 1421-1427.

[10] WANG J, LI T F, LU H B, et al. Penalized weighted least-squares approach to sinogram noise reduction and image reconstruction for 1ow-dose X-ray computed tomography. IEEE Transactions on Medical Imaging, 2006, 25(10): 1272-1283.

[11] LI T F, LI X, WANG J. Nonlinear sinogram smoothing for 1ow-dose X-Ray CT. IEEE Transactions on Nuclear Science, 2004, 51(5): 2505-2513.

[12] ZHANG W K, LI J S, SUN J Q, et al. FBP initialized few-view CT reconstruction algorithm using similar prior image constraint//38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, August 16-20, 2016, Orlando, FL, USA. New York: IEEE, 2016: 3949-3952.

[13] FANG L, LI L. Application progress of adaptive statistical iterative reconstruction in reducing radiation dose. CT Theory and Applications, 2013, 22(2): 207-213.

[14] SHIN H B, KIM M S, LAW M, et al. Application of sigmoidal optimization to reconstruct nuclear medicine image: Comparison with filtered back projection and iterative reconstruction method. Nuclear Engineering and Technology, 2021, 53(1): 258-265.

[15] ZHANG Y K, ZHANG J Y, LU H B. Statistical sinogram smoothing for low-dose CT with segmentation-based adaptive filtering. IEEE Transactions on Nuclear Science, 2010, 57(5): 2587-2598.

[16] ZHANG Q, GUI Z, CHEN Y, et al. Bayesian sinogram smoothing with an anisotropic diffusion weighted prior for low-dose X-ray computed tomography. Optik, 2013, 124(17): 2811-2816.

[17] ELBAKRI I A, FESSLER J A. Efficient and accurate likelihood for iterative image reconstruction in X-ray computed tomography//Medical Imaging 2003, June 20, 2003, San Diego, CA, USA. Bellingham: SPIE, 2003: 1839-1850.

[18] CHEN Y,YANG Z, HU Y, et al. Thoracic low-dose CT image processing using an artifact suppressed large-scale nonlocal means. Physics in Medicine and Biology, 2013, 57(9): 2667-2688.

[19] YIN J, FAN Y L, QIN S H. Research of CT reconstruction algorithms based on compressed sensing. China Medical Devices, 2019, 34(8): 166-169.

[20] ZHANG K, ZUO W M, CHEN Y J, et al. Beyond a Gaussian denoiser: residual learning of deep CNN for image denoising. IEEE Transactions on Image Processing: a Publication of the IEEE Signal Processing Society, 2017, 26(7): 3142-3155.

[21] HAN Y S, YOO J, YE J C. Deep residual learning for compressed sensing CT Reconstruction via persistent homology analysis. arXiv:1611.06391.

[22] LEDIG C, THEIS L, HUSZÁR F, et al. Photo-realistic single image super-resolution using a generative adversarial network//IEEE Conference on Computer Vision and Pattern Recognition, July 21-26, 2017, Honolulu, HI, USA. Piscataway: IEEE, 2017: 4681-4690.

[23] DONG C, LOY C C, HE K M, et al. Image super-resolution using deep convolutional networks. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2016, 38(2): 295-307.

[24] WANG K F, GOU C, DUAN Y J, et al. Generative adversarial networks: the state of the art and beyond. Acta Automatica Sinica, 2017, 43(3): 321-332.

[25] WANG G M, QIAO J F, WANG L. A generative adversarial network based on energy function. Acta Automatica Sinica, 2018, 44(5): 28-38.

[26] CHEN J W, CHEN J W, CHAO H Y, et al. Image blind denoising with generative adversarial network based noise modeling//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, June 18-23, 2018, Salt Lake City, UT, USA. Piscataway: IEEE, 2018: 3155-3164.

[27] HOWARD A, SANDLER M, CHEN B, et al. Searching for MobileNetV3//2019 IEEE/CVF International Conference on Computer Vision,October 27-November 2, 2019, Seoul, Korea (South). Piscataway: IEEE, 2019: 1314-1324.