Abstract
Self-supervised image denoising techniques emerged as convenient methods that allow training denoising models without requiring ground-truth noise-free data. Existing methods usually optimize loss metrics that are calculated from multiple noisy realizations of similar images, e.g., from neighboring tomographic slices. However, those approaches fail to utilize the multiple contrasts that are routinely acquired in medical imaging modalities like MRI or dual-energy CT. In this work, we propose the new self-supervised training scheme Noise2Contrast that combines information from multiple measured image contrasts to train a denoising model. We stack denoising with domain-transfer operators to utilize the independent noise realizations of different image contrasts to derive a self-supervised loss. The trained denoising operator achieves convincing quantitative and qualitative results, outperforming state-of-the-art self-supervised methods by 4.7–11.0%/4.8–7.3% (PSNR/SSIM) on brain MRI data and by 43.6–50.5%/57.1–77.1% (PSNR/SSIM) on dual-energy CT X-ray microscopy data with respect to the noisy baseline. Our experiments on different real measured data sets indicate that Noise2Contrast training generalizes to other multi-contrast imaging modalities.
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
References
Batson, J., Royer, L.: Noise2Self: blind denoising by self-supervision. In: Proceedings of the ICML, pp. 524–533. PMLR (2019)
Choi, K., Lim, J.S., Kim, S.: Self-supervised inter-and intra-slice correlation learning for low-dose CT image restoration without ground truth. Expert Syst. Appl. 209, 118072 (2022)
Denck, J., Guehring, J., Maier, A., Rothgang, E.: Enhanced magnetic resonance image synthesis with contrast-aware generative adversarial networks. J. Imaging 7(8), 133 (2021)
Genant, H.K., Boyd, D.: Quantitative bone mineral analysis using dual energy computed tomography. Investig. Radiol. 12(6), 545–551 (1977)
Jeon, S.Y., Kim, W., Choi, J.H.: MM-net: multi-frame and multi-mask-based unsupervised deep denoising for low-dose computed tomography. IEEE TRPMS 1–12 (2022)
Kim, K., Kwon, T., Ye, J.C.: Noise distribution adaptive self-supervised image denoising using tweedie distribution and score matching. In: Proceedings of the CVPR, pp. 2008–2016 (2022)
Krull, A., Buchholz, T.O., Jug, F.: Noise2Void-learning denoising from single noisy images. In: Proceedings of the CVPR, pp. 2129–2137 (2019)
Lehtinen, J., et al.: Noise2Noise: learning image restoration without clean data. In: Proceedings of the PMLR, vol. 80, pp. 2965–2974. PMLR (2018)
Maier, A.K., et al.: Learning with known operators reduces maximum error bounds. Nat. Mach. Intell. 1(8), 373–380 (2019)
Prah, D.E., Paulson, E.S., Nencka, A.S., Schmainda, K.M.: A simple method for rectified noise floor suppression: phase-corrected real data reconstruction with application to diffusion-weighted imaging. Magn. Reson. Med. 64(2), 418–429 (2010)
Ronneberger, O., Fischer, P., Brox, T.: U-net: convolutional networks for biomedical image segmentation. In: Navab, N., Hornegger, J., Wells, W.M., Frangi, A.F. (eds.) MICCAI 2015. LNCS, vol. 9351, pp. 234–241. Springer, Cham (2015). https://doi.org/10.1007/978-3-319-24574-4_28
Schmainda, K.M., Prah, M.A.: Data from brain-tumor-progression. Technical report Version 1, The Cancer Imaging Archive (2018). https://doi.org/10.7937/K9/TCIA.2018.15quzvnb
Thies, M., et al.: Calibration by differentiation-self-supervised calibration for X-ray microscopy using a differentiable cone-beam reconstruction operator. J. Microsc. 287(2), 81–92 (2022)
Tomasi, C., Manduchi, R.: Bilateral filtering for gray and color images. In: Proceedings of the ICCV, pp. 839–846. IEEE (1998)
Wagner, F., et al.: Trainable joint bilateral filters for enhanced prediction stability in low-dose CT. Sci. Rep. 12(1), 1–9 (2022)
Wagner, F., et al.: Ultralow-parameter denoising: trainable bilateral filter layers in computed tomography. Med. Phys. 49(8), 5107–5120 (2022)
Wagner, F., et al.: Monte Carlo dose simulation for in-vivo X-ray nanoscopy. In: Maier-Hein, K., Deserno, T.M., Handels, H., Maier, A., Palm, C., Tolxdorff, T. (eds.) Bildverarbeitung für die Medizin 2022. Informatik aktuell, pp. 107–112. Springer, Wiesbaden (2022). https://doi.org/10.1007/978-3-658-36932-3_22
Wu, D., Ren, H., Li, Q.: Self-supervised dynamic CT perfusion image denoising with deep neural networks. IEEE Trans. Radiat. Plasma Med. Sci. 5(3), 350–361 (2020)
Zhang, Z., Liang, X., Zhao, W., Xing, L.: Noise2Context: context-assisted learning 3D thin-layer for low-dose CT. Med. Phys. 48(10), 5794–5803 (2021)
Acknowledgements
This work was supported by the European Research Council (ERC Grant No. 810316) and a GPU donation through the NVIDIA Hardware Grant Program.
F.W. conceived and conducted the experiments. M.T., L.P., N.M., M.G., J.U., and J.-H.C. provided valuable technical feedback during development. S.P., O.A., D.W., G.N., and S.U. prepared and scanned the bone samples. A.M. supervised the project. All authors reviewed the manuscript. L.P. and N.M. are employees of Siemens Healthcare GmbH.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Wagner, F. et al. (2023). Noise2Contrast: Multi-contrast Fusion Enables Self-supervised Tomographic Image Denoising. In: Frangi, A., de Bruijne, M., Wassermann, D., Navab, N. (eds) Information Processing in Medical Imaging. IPMI 2023. Lecture Notes in Computer Science, vol 13939. Springer, Cham. https://doi.org/10.1007/978-3-031-34048-2_59
Download citation
DOI: https://doi.org/10.1007/978-3-031-34048-2_59
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-34047-5
Online ISBN: 978-3-031-34048-2
eBook Packages: Computer ScienceComputer Science (R0)