Abstract
The recent achievements of Deep Learning rely on the test data being similar in distribution to the training data. In an ideal case, Deep Learning models would achieve Out-of-Distribution (OoD) Generalization, i.e. reliably make predictions on out-of-distribution data. Yet in practice, models usually fail to generalize well when facing a shift in distribution. Several methods were thereby designed to improve the robustness of the features learned by a model through Regularization- or Domain-Prediction-based schemes. Segmenting medical images such as MRIs of the hippocampus is essential for the diagnosis and treatment of neuropsychiatric disorders. But these brain images often suffer from distribution shift due to the patient’s age and various pathologies affecting the shape of the organ. In this work, we evaluate OoD Generalization solutions for the problem of hippocampus segmentation in MR data using both fully- and semi-supervised training. We find that no method performs reliably in all experiments. Only the V-REx loss stands out as it remains easy to tune, while it outperforms a standard U-Net in most cases.
Supported by the Bundesministerium für Gesundheit (BMG) with grant [ZMVI1-2520DAT03A]. The final authenticated version of this manuscript will be published in Lecture Notes in Pattern recognition in the life and natural sciences.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Ahuja, K., Shanmugam, K., Varshney, K.R., Dhurandhar, A.: Invariant risk minimization games (2020). http://arxiv.org/abs/2002.04692
Arjovsky, M., Bottou, L., Gulrajani, I., Lopez-Paz, D.: Invariant risk minimization. http://arxiv.org/abs/1907.02893
Boccardi, M., et al.: Training labels for hippocampal segmentation based on the EADC-ADNI harmonized hippocampal protocol 11(2), 175–183 (2015). https://doi.org/10.1016/j.jalz.2014.12.002, https://linkinghub.elsevier.com/retrieve/pii/S155252601402891X
Carmo, D., Silva, B., Yasuda, C., Rittner, L., Lotufo, R.: Hippocampus segmentation on epilepsy and Alzheimer’s disease studies with multiple convolutional neural networks (2020). http://arxiv.org/abs/2001.05058
Castro, D.C., Walker, I., Glocker, B.: Causality matters in medical imaging. Nat. Commun. 11(1) (2020). https://doi.org/10.1038/s41467-020-17478-w
Dinsdale, N.K., Jenkinson, M., Namburete, A.I.L.: Deep learning-based unlearning of dataset bias for MRI harmonisation and confound removal. bioRxiv (2020). https://doi.org/10.1101/2020.10.09.332973, https://www.biorxiv.org/content/early/2020/12/14/2020.10.09.332973
Ganin, Y., Lempitsky, V.: Unsupervised domain adaptation by backpropagation (2015). http://arxiv.org/abs/1409.7495
Isensee, F., et al.: nnU-net: self-adapting framework for U-net-based medical image segmentation (2018). http://arxiv.org/abs/1809.10486
Krueger, D., et al.: Out-of-distribution generalization via risk extrapolation (REx) (2020). http://arxiv.org/abs/2003.00688
Kulaga-Yoskovitz, J., et al.: Multi-contrast submillimetric 3 tesla hippocampal subfield segmentation protocol and dataset 2(1), 150059 (2015). https://doi.org/10.1038/sdata.2015.59, http://www.nature.com/articles/sdata201559
Litjens, G., et al.: A survey on deep learning in medical image analysis. Med. Image Anal. 42 (2017). https://doi.org/10.1016/j.media.2017.07.005
Simpson, A.L., et al.: A large annotated medical image dataset for the development and evaluation of segmentation algorithms (2019). http://arxiv.org/abs/1902.09063
Xu, Y., et al.: Age effects on hippocampal structural changes in old men: the HAAS. NeuroImage 40(3), 1003–1015 (2008) https://doi.org/10.1016/j.neuroimage.2007.12.034, https://www.sciencedirect.com/science/article/pii/S105381190701141X
Xue, Y., Feng, S., Zhang, Y., Zhang, X., Wang, Y.: Dual-task self-supervision for cross-modality domain adaptation. In: Martel, A.L., et al. (eds.) MICCAI 2020. LNCS, vol. 12261, pp. 408–417. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-59710-8_40
Zhu, H., et al.: Dilated dense u-net for infant hippocampus subfield segmentation 13, 30 (2019) https://doi.org/10.3389/fninf.2019.00030, https://www.frontiersin.org/article/10.3389/fninf.2019.00030/full
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Sanner, A., González, C., Mukhopadhyay, A. (2021). How Reliable Are Out-of-Distribution Generalization Methods for Medical Image Segmentation?. In: Bauckhage, C., Gall, J., Schwing, A. (eds) Pattern Recognition. DAGM GCPR 2021. Lecture Notes in Computer Science(), vol 13024. Springer, Cham. https://doi.org/10.1007/978-3-030-92659-5_39
Download citation
DOI: https://doi.org/10.1007/978-3-030-92659-5_39
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-92658-8
Online ISBN: 978-3-030-92659-5
eBook Packages: Computer ScienceComputer Science (R0)