Abstract
Unsupervised anomaly detection in medical imaging aims to detect and localize arbitrary anomalies without requiring labels during training. Often, this is achieved by learning a data distribution of normal samples and detecting anomalies as regions in the image which deviate from this distribution. In the medical imaging domain, most current state-of-the-art methods use latent variable generative models operating directly on the images. However, generative models have been shown to mostly capture low-level features s.a. pixel-intensities instead of rich semantic features, which also applies to their representations. We circumvent this problem by proposing CRADL whose core idea is to model the distribution of normal samples directly in the low-dimensional representation space of an encoder which has been trained with a contrastive pretext-task. By utilizing the representations of contrastive learning we aim to fix the over-fixation on low-level features and aim to learn more semantic-rich representations. Our experiments on the task of anomaly localization on three distinct datasets show that 1) the contrastive representations are superior to generative latent variable models and 2) the CRADL framework shows competetive or superior performance to state-of-the-art.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Baur C, Denner S, Wiestler B, Navab N, Albarqouni S. Autoencoders for unsupervised anomaly segmentation in brain MR images: a comparative study. Med Image Anal. 2021;69:101952.
Chen X,You S, TezcanKC,Konukoglu E.Unsupervised lesion detection via image restoration with a normative prior. Med Image Anal. 2020;64:101713.
Schlegl T, Seeböck P,Waldstein SM, Langs G, Schmidt-Erfurth U. F-AnoGAN: fast unsupervised anomaly detection with generative adversarial networks. Med Image Anal. 2019;54:30– 44.
Nalisnick E, Matsukawa A, Teh YW, Gorur D, Lakshminarayanan B. Do deep generative models know what they don’t know? arXiv:1810.09136. 2019.
Ren J, Liu PJ, Fertig E, Snoek J, Poplin R, DePristo MA et al. Likelihood ratios for out-ofdistribution detection. arXiv:1906.02845. 2019.
Xiao Z, Yan Q, Amit Y. Likelihood regret: an out-of-distribution detection score for variational auto-encoder. arXiv:2003.02977. 2020.
Meissen F, Kaissis G, Rueckert D. Challenging current semi-supervised anomaly segmentation methods for brain MRI. 2021.
Zimmerer D, Kohl SAA, Petersen J, Isensee F, Maier-Hein KH. Context-encoding variational autoencoder for unsupervised anomaly detection. arXiv:1812.05941. 2018.
Chen T, Kornblith S, Norouzi M, Hinton G. A simple framework for contrastive learning of visual representations. arXiv: 2002.05709. 2020.
Dempster AP, Laird NM, Rubin DB. Maximum Likelihood from incomplete data via the EM Algorithm. Journal of the Royal Statistical Society: Series B (Methodological). 1977;39(1):1– 22.
Van Essen D, Ugurbil K, Auerbach E, Barch D, Behrens T, Bucholz R et al. The human connectome project: a data acquisition perspective. NeuroImage. 2012;62(4):2222–31.
Zimmerer D, Petersen J, Köhler G, Jäger P, Full P, Roß T et al. Medical out-of-distribution analysis challenge. 2020.
Bakas S, Akbari H, Sotiras A, Bilello M, Rozycki M, Kirby JS et al. Advancing the Cancer GenomeAtlas GliomaMRIcollections with expert segmentation labels and radiomic features. Sci Data. 2017;4(1):170117.
Maier O, Menze BH, von der Gablentz J, Häni L, Heinrich MP, LiebrandMet al. ISLES 2015 - a public evaluation benchmark for ischemic stroke lesion segmentation from multispectral MRI. Med Image Anal. 2017;35:250–69.
Radford A, Metz L, Chintala S.Unsupervised representation learning with deep convolutional generative adversarial networks. arXiv:1511.06434. 2016.
Loshchilov I, Hutter F. SGDR: stochastic gradient descent with warm restarts. arXiv: 1608.03983. 2017.
Zimmerer D, Isensee F, Petersen J,Kohl S, Maier-Hein K.Unsupervised anomaly localization using variational auto-encoders. arXiv:1907.02796. 2019.
Li CL, Sohn K, Yoon J, Pfister T. CutPaste: self-supervised learning for anomaly detection and localization. CVPR. 2021:9659–69.
He K, Chen X, Xie S, Li Y, Dollár P, Girshick R. Masked autoencoders are scalable vision learners. 2021.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 Der/die Autor(en), exklusiv lizenziert an Springer Fachmedien Wiesbaden GmbH, ein Teil von Springer Nature
About this paper
Cite this paper
Lüth, C.T., Zimmerer, D., Koehler, G., Jaeger, P.F., Isenensee, F., Maier-Hein, K.H. (2023). Contrastive Representations for Unsupervised Anomaly Detection and Localization. In: Deserno, T.M., Handels, H., Maier, A., Maier-Hein, K., Palm, C., Tolxdorff, T. (eds) Bildverarbeitung für die Medizin 2023. BVM 2023. Informatik aktuell. Springer Vieweg, Wiesbaden. https://doi.org/10.1007/978-3-658-41657-7_54
Download citation
DOI: https://doi.org/10.1007/978-3-658-41657-7_54
Published:
Publisher Name: Springer Vieweg, Wiesbaden
Print ISBN: 978-3-658-41656-0
Online ISBN: 978-3-658-41657-7
eBook Packages: Computer Science and Engineering (German Language)