Abstract
The acquisition conditions for low-dose and high-dose CT images are usually different, so that the shifts in the CT numbers often occur. Accordingly, unsupervised deep learning-based approaches, which learn the target image distribution, often introduce CT number distortions and result in detrimental effects in diagnostic performance. To address this, here we propose a novel unsupervised learning approach for lowdose CT reconstruction using patch-wise deep metric learning. The key idea is to learn embedding space by pulling the positive pairs of image patches which shares the same anatomical structure, and pushing the negative pairs which have same noise level each other. Thereby, the network is trained to suppress the noise level, while retaining the original global CT number distributions even after the image translation. Experimental results confirm that our deep metric learning plays a critical role in producing high quality denoised images without CT number shift.
This study was approved by the institutional review board (IRB) of our institution, and the requirement for informed consent was waived. This study was supported by the National Research Foundation of Korea (NRF) grant NRF-2020R1A2B5B03001980, KAIST Key Research Institute (Interdisciplinary Research Group) Project, Field-oriented Technology Development Project for Customs Administration through NRF funded by the Ministry of Science & ICT (MSIT) and Korea Customs Service (NRF-2021M3I1A1097938). Sunkyoung You was supported by the Bio & Medical Technology Development Program of the NRF & funded by the MSIT (No. NRF-2019M3E5D1A02068564)
Joonhyung Lee is currently at VUNO Corp.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Ahn, S.H., Kim, Y.W., Baik, S.K., Hwang, Yeon, J., Lee, I.W.: Usefulness of computed tomography hounsfield unit measurement for diagnosis of congenital cholesteatoma. J. Korean Radiol. Soc. 70, 153–158 (2014)
Bai, T., Wang, B., Nguyen, D., Jiang, S.: Probabilistic self-learning framework for low-dose CT denoising. Med. Phys. 48(5), 2258–2270 (2021)
Brenner, D.J., Hall, E.J.: Computed tomography - An increasing source of radiation exposure, November 2007. https://doi.org/10.1056/NEJMra072149. http://www.nejm.org/doi/abs/10.1056/NEJMra072149
Chen, H., et al.: Low-dose CT With a residual encoder-decoder convolutional neural network. IEEE Trans. Med. Imaging 36(12), 2524–2535 (2017)
Thukral, C.L., Singh, A., Singh, S., Sood, A.S., Singh, K.: Role of high resolution computed tomography in evaluation of pathologies of temporal bone. J. Clin. Diagnostic Res. 9, TC07–TC10 (2015)
Cruz-Bastida, J.P., Zhang, R., Gomez-Cardona, D., Hayes, J., Li, K., Chen, G.H.: Impact of noise reduction schemes on quantitative accuracy of ct numbers. Med. Phys. 46(7), 3013–3024 (2019)
Gu, J., Yang, T.S., Ye, J.C., Yang, D.H.: CycleGAN denoising of extreme low-dose cardiac CT using wavelet-assisted noise disentanglement. Med. Image Anal. 74, 102209 (2021)
Gu, J., Ye, J.C.: AdaIN-based tunable cycleGAN for efficient unsupervised low-dose CT denoising. IEEE Trans. Comput. Imaging 7, 73–85 (2021)
Healthcare, G.: TrueFidelity CT. Technical white paper on deep learning image reconstruction. https://www.gehealthcare.co.kr/products/computed-tomography/truefidelity
Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization. In: Bengio, Y., LeCun, Y. (eds.) 3rd International Conference on Learning Representations, (ICLR) 2015(May), pp. 7–9, 2015. Conference Track Proceedings. San Diego, CA, USA (2015). http://arxiv.org/abs/1412.6980
Kwon, T., Ye, J.C.: Cycle-free cyclegan using invertible generator for unsupervised low-dose CT denoising. IEEE Trans. Comput. Imaging 7, 1354–1368 (2021)
Mao, X., Li, Q., Xie, H., Lau, R.Y., Wang, Z., Smolley, S.P.: Least squares generative adversarial networks. In: Proceedings of the IEEE International Conference on Computer Vision. vol. 2017-October, pp. 2813–2821. Institute of Electrical and Electronics Engineers Inc., December 2017. https://doi.org/10.1109/ICCV.2017.304. https://ieeexplore.ieee.org/document/8237566
Park, M.H., Rah, Y.C., Kim, Y.H., hoon Kim, J.: Usefulness of computed tomography hounsfield unit density in preoperative detection of cholesteatoma in mastoid ad antrum. Am. J. Otolaryngology 32(3), 194–197 (2011)
Park, T., Efros, A.A., Zhang, R., Zhu, J.-Y.: Contrastive Learning for Unpaired Image-to-Image Translation. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, J.-M. (eds.) ECCV 2020. LNCS, vol. 12354, pp. 319–345. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58545-7_19
Paszke, A., et al.: PyTorch: an imperative style, high-performance deep learning library. In: Wallach, H., Larochelle, H., Beygelzimer, A., d’Alché Buc, F., Fox, E., Garnett, R. (eds.) Advances in Neural Information Processing Systems 32, pp. 8026–8037. Curran Associates, Inc. (2019)
Sim, B., Oh, G., Kim, J., Jung, C., Ye, J.C.: Optimal transport driven cyclegan for unsupervised learning in inverse problems. SIAM J. Imag. Sci. 13(4), 2281–2306 (2020)
Tang, C., et al.: Unpaired low-dose CT denoising network based on cycle-consistent generative adversarial network with prior image information. Computational and Mathematical Methods in Medicine 2019 (2019)
Wang, W., Zhou, W., Bao, J., Chen, D., Li, H.: Instance-wise hard negative example generation for contrastive learning in unpaired image-to-image translation. In: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), pp. 14020–14029, October 2021
Wang, X., et al.: ESRGAN: enhanced super-resolution generative adversarial networks. In: The European Conference on Computer Vision Workshops (ECCVW) (2018)
Yang, Q., et al.: Low-Dose CT image denoising using a generative adversarial network with wasserstein distance and perceptual loss. IEEE Trans. Med. Imaging 37(6), 1348–1357 (2018). https://doi.org/10.1109/TMI.2018.2827462
You, C., et al.: CT super-resolution GAN constrained by the identical, residual, and cycle learning ensemble (GAN-circle). IEEE Trans. Med. Imaging 39(1), 188–203 (2020)
Yuan, N., Zhou, J., Qi, J.: Half2half: deep neural network based CT image denoising without independent reference data 65(21), 215020, November 2020. https://doi.org/10.1088/1361-6560/aba939. https://doi.org/10.1088/1361-6560/aba939
Zhu, J.Y., Park, T., Isola, P., Efros, A.A.: Unpaired image-to-image translation using cycle-consistent adversarial networks. In: Proceedings of the IEEE International Conference on Computer Vision, vol. 2017, pp. 2242–2251. Institute of Electrical and Electronics Engineers Inc., December 2017. https://doi.org/10.1109/ICCV.2017.244
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
1 Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Jung, C., Lee, J., You, S., Ye, J.C. (2022). Patch-Wise Deep Metric Learning for Unsupervised Low-Dose CT Denoising. In: Wang, L., Dou, Q., Fletcher, P.T., Speidel, S., Li, S. (eds) Medical Image Computing and Computer Assisted Intervention – MICCAI 2022. MICCAI 2022. Lecture Notes in Computer Science, vol 13436. Springer, Cham. https://doi.org/10.1007/978-3-031-16446-0_60
Download citation
DOI: https://doi.org/10.1007/978-3-031-16446-0_60
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-16445-3
Online ISBN: 978-3-031-16446-0
eBook Packages: Computer ScienceComputer Science (R0)