Abstract
The scarce availability of labeled data makes multi-modal domain adaptation an interesting approach in medical image analysis. Deep learning-based registration methods, however, still struggle to outperform their non-trained counterparts. Supervised domain adaptation also requires labeled- or other ground truth data. Hence, unsupervised domain adaptation is a valuable goal, that has so far mainly shown success in classification tasks. We are the first to report unsupervised domain adaptation for discrete displacement registration using classifier discrepancy in medical imaging. We train our model with mono-modal registration supervision. For cross-modal registration no supervision is required, instead we use the discrepancy between two classifiers as training loss. We also present a new projected Earth Mover's distance for measuring classifier discrepancy. By projecting the 2D distributions to 1D histograms, the EMD L1 distance can be computed using their cumulative sums.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Xiao Y, Rivaz H, Chabanas M, et al. Evaluation of MRI to ultrasound registration methods for brain shift correction: the CuRIOUS2018 challenge. IEEE Trans Med Imaging. 2020 mar;39(3):777–786.
Hu Y, Modat M, Gibson E, et al. Weakly-supervised convolutional neural networks for multimodal image registration. Med Image Anal. 2018;49:1–13. Available from: https://www.sciencedirect.com/science/article/pii/S1361841518301051.
Lee CY, Batra T, Baig MH, et al. Sliced wasserstein discrepancy for unsupervised domain adaptation. Procs IEEE CVPR. 2019; p. 10277–10287. Available from: https://arxiv.org/abs/1903.04064.
Saito K, Watanabe K, Ushiku Y, et al. Maximum classifier discrepancy for unsupervised domain adaptation. Procs IEEE CVPR. 2018; p. 3723–3732. Available from: https://arxiv.org/abs/1712.02560.
Werman M, Peleg S, Rosenfeld A. A distance metric for multidimensional histograms. Comp Vis Graph Image Proc. 1985;32(3):328–336. Available from: https://www.sciencedirect.com/science/article/pii/0734189X85900556.
Haibin Ling, Okada K. Diffusion distance for histogram comparison. Procs IEEE CVPR. 2006 jun;1:246–253.
Heinrich MP, Hansen L. Unsupervised learning of multimodal image registration using domain adaptation with projected earth move's discrepancies. Medical Imaging with Deep Learning. 2020 may;Available from: https://arxiv.org/abs/2005.14107.
Simonovsky M, Gutiérrez-Becker B, Mateus D, et al. A deep metric for multimodal registration. Proc MICCAI. 2016; p. 10–18.
Hering A, Kuckertz S, Heldmann S, et al. Memory-efficient 2.5D convolutional transformer networks for multi-modal deformable registration with weak label supervision applied to whole-heart CT and MRI scans. Int J Comput Assist Radiol Surg. 2019;14(11):1901–1912. Available from: https://doi.org/10.1007/s11548-019-02068-z.
Heinrich MP. Closing the gap between deep and conventional image registration using probabilistic dense displacement networks. Lect Notes Comp Scie. 2019;11769:50–58. Available from: https://arxiv.org/abs/1907.10931.
Wolterink JM, Dinkla AM, Savenije MHF, et al. Deep MR to CT synthesis using unpaired data. In: Tsaftaris SA, Gooya A, Frangi AF, et al., editors. Simulation and Synthesis in Medical Imaging. Cham: Springer International Publishing; 2017. p. 14–23.
Dou Q, Liu Q, Heng PA, et al. Unpaired multi-modal segmentation via knowledge distillation. IEEE Trans Med Imaging. 2020;39(7):2415–2425.
Heinrich MP, Oktay O, Bouteldja N. OBELISK – one kernel to solve nearly everything: unified 3D binary convolutions for image analysis. In: MidlProceedings of the Conference on Medical Imaging with Deep Learning (MIDL). Amsterdam; 2018. p. 9.
Hirschmuller H. Accurate and efficient stereo processing by semi-global matching and mutual information. Proc IEEE CVPR. 2005; p. 807–814.
Jimenez-del Toro O, Müller H, Krenn M, et al. Cloud-based evaluation of anatomical structure segmentation and landmark detection algorithms: VISCERAL anatomy benchmarks. IEEE Trans Med Imaging. 2016 nov;35(11):2459–2475.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Der/die Autor(en), exklusiv lizenziert durch Springer Fachmedien Wiesbaden GmbH, ein Teil von Springer Nature
About this paper
Cite this paper
Kruse, C.N., Hansen, L., Heinrich, M.P. (2021). Multi-modal Unsupervised Domain Adaptation for Deformable Registration Based on Maximum Classifier Discrepancy. In: Palm, C., Deserno, T.M., Handels, H., Maier, A., Maier-Hein, K., Tolxdorff, T. (eds) Bildverarbeitung für die Medizin 2021. Informatik aktuell. Springer Vieweg, Wiesbaden. https://doi.org/10.1007/978-3-658-33198-6_47
Download citation
DOI: https://doi.org/10.1007/978-3-658-33198-6_47
Published:
Publisher Name: Springer Vieweg, Wiesbaden
Print ISBN: 978-3-658-33197-9
Online ISBN: 978-3-658-33198-6
eBook Packages: Computer Science and Engineering (German Language)