Abstract
Anomaly detection in medical imaging is a challenging task in contexts where abnormalities are not annotated. This problem can be addressed through unsupervised anomaly detection (UAD) methods, which identify features that do not match with a reference model of normal profiles. Artificial neural networks have been extensively used for UAD but they do not generally achieve an optimal trade-off between accuracy and computational demand. As an alternative, we investigate mixtures of probability distributions whose versatility has been widely recognized for a variety of data and tasks, while not requiring excessive design effort or tuning. Their expressivity makes them good candidates to account for complex multivariate reference models. Their much smaller number of parameters makes them more amenable to interpretation and efficient learning. However, standard estimation procedures, such as the Expectation-Maximization algorithm, do not scale well to large data volumes as they require high memory usage. To address this issue, we propose to incrementally compute inferential quantities. This online approach is illustrated on the challenging detection of subtle abnormalities in MR brain scans for the follow-up of newly diagnosed Parkinsonian patients. The identified structural abnormalities are consistent with the disease progression, as accounted by the Hoehn and Yahr scale.
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
Arnaud, A., Forbes, F., Coquery, N., Collomb, N., Lemasson, B., Barbier, E.: Fully automatic lesion localization and characterization: application to brain tumors using multiparametric quantitative MRI data. IEEE Trans. Med. Imaging 37(7), 1678–1689 (2018)
Baudry, J.P., Maugis, C., Michel, B.: Slope heuristic: overview and implementation. Stat. Comp. 22, 455–470 (2012)
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. 69, 101952 (2021)
Borkar, V.: Stochastic Approximation: A Dynamical View Point. Cambridge University Press (2008)
Bourdon, A., Noureddine, A., Rouvoy, R., Seinturier, L.: PowerAPI: a software library to monitor the energy consumed at the process-level. ERCIM News (2013)
Cappé, O., Moulines, E.: On-line Expectation-Maximization algorithm for latent data models. J. R. Stat. Soc. B 71, 593–613 (2009)
Dexter, D.T., et al.: Increased nigral iron content in postmortem Parkinsonian brain. Lancet 2, 1219–1220 (1987)
Du, G., et al.: Combined R2* and diffusion tensor imaging changes in the substantia Nigra in Parkinson’s disease. Mov. Disord. 26(9), 1627–1632 (2011)
Evchenko, M., Vanschoren, J., Hoos, H.H., Schoenauer, M., Sebag, M.: Frugal machine learning. arXiv arXiv:abs/2111.03731 (2021)
Forbes, F., Wraith, D.: A new family of multivariate heavy-tailed distributions with variable marginal amounts of tailweights: application to robust clustering. Stat. Comput. 24(6), 971–984 (2014)
Fort, G., Moulines, E., Wai, H.T.: A stochastic path-integrated differential estimator expectation maximization algorithm. In: 34th Conference on Neural Information Processing Systems (NeurIPS) (2020)
Fort, G., Gach, P., Moulines, E.: Fast incremental expectation maximization for finite-sum optimization: nonasymptotic convergence. Stat. Comp. 31, 48 (2021)
Garyfallidis, E., et al.: Dipy, a library for the analysis of diffusion MRI data. Front, Neuroinf. 8, 1–17 (2014)
Gillies, R.J., Kinahan, P.E., Hricak, H.: Radiomics: images are more than pictures, they are data. Radiology 278(2), 563–77 (2016)
He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 770–778 (2016)
Hoehn, M.M., Yahr, M.D.: Parkinsonism: onset, progression, and mortality. Neurology 50(2), 318–318 (1998)
Karimi, B., Miasojedow, B., Moulines, E., Wai, H.T.: Non-asymptotic analysis of biased stochastic approximation scheme. Proc. Mach. Learn. Res. 99, 1–31 (2019)
Karimi, B., Wai, H.T., Moulines, E., Lavielle, M.: On the global convergence of (fast) incremental Expectation Maximization methods. In: 33rd Conference on Neural Information Processing Systems (NeurIPS) (2019)
Kotz, S., Nadarajah, S.: Multivariate t Distributions And Their Applications. Cambridge University Press (2004)
Kuhn, E., Matias, C., Rebafka, T.: Properties of the stochastic approximation EM algorithm with mini-batch sampling. Stat. Comput. 30(6), 1725–1739 (2020). https://doi.org/10.1007/s11222-020-09968-0
Lagogiannis, I., Meissen, F., Kaissis, G., Rueckert, D.: Unsupervised pathology detection: A deep dive into the state of the art. arXiv arXiv:abs:2303.00609 (2023)
Li, C., Sohn, K., Yoon, J., Pfister, T.: CutPaste: self-supervised learning for anomaly detection and localization. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 9664–9674 (2021)
Maire, F., Moulines, E., Lefebvre, S.: Online EM for functional data. Comput. Stat. Data Anal. 111, 27–47 (2017)
Marek, K., et al.: The Parkinson’s progression markers initiative - establishing a PD biomarker cohort. Ann. Clin. Transl. Neurol. 5, 1460–1477 (2018)
McLachlan, G.J., Krishnan, T.: The EM Algorithm and Extensions. Wiley (2007)
Munoz-Ramirez, V., Forbes, F., Arbel, J., Arnaud, A., Dojat, M.: Quantitative MRI characterization of brain abnormalities in de novo Parkinsonian patients. In: IEEE International Symposium on Biomedical Imaging (2019)
Muñoz-Ramírez, V., Kmetzsch, V., Forbes, F., Meoni, S., Moro, E., Dojat, M.: Subtle anomaly detection in MRI brain scans: application to biomarkers extraction in patients with de novo Parkinson’s disease. Artif. Intell. Med. 125, 102251 (2021)
Muñoz-Ramírez, V., Pinon, N., Forbes, F., Lartizen, C., Dojat, M.: Patch vs. global image-based unsupervised anomaly detection in MR brain scans of early Parkinsonian Patients. In: Machine Learning in Clinical Neuroimaging (2021)
Nguyen, H.D., Forbes, F.: Global implicit function theorems and the online expectation-maximisation algorithm. Aust. NZ J. Stat. 64, 255–281 (2022)
Nguyen, H.D., Forbes, F., McLachlan, G.J.: Mini-batch learning of exponential family finite mixture models. Stat. Comput. 30(4), 731–748 (2020). https://doi.org/10.1007/s11222-019-09919-4
Oluwasegun, A., Jung, J.C.: A multivariate Gaussian mixture model for anomaly detection in transient current signature of control element drive mechanism. Nucl. Eng. Des. 402, 112098 (2023)
Pauli, W.M., Nili, A.N., Tyszka, J.M.: A high-resolution probabilistic in vivo atlas of human subcortical brain nuclei. Sci. Data 5(1), 1–13 (2018)
Pinaya, W.H., et al.: Unsupervised brain imaging 3D anomaly detection and segmentation with transformers. Med. Image Anal. 79, 102475 (2022)
Pinon, N., Oudoumanessah, G., Trombetta, R., Dojat, M., Forbes, F., Lartizien, C.: Brain subtle anomaly detection based on auto-encoders latent space analysis: application to de novo Parkinson patients. In: IEEE International Symposium on Biomedical Imaging (2023)
Schwartz, R., Dodge, J., Smith, N., Etzioni, O.: Green AI. Commun. ACM 63(12), 54–63 (2020)
Schwarz, S.T., Abaei, M., Gontu, V., Morgan, P.S., Bajaj, N., Auer, D.P.: Diffusion tensor imaging of nigral degeneration in Parkinson’s disease: a region-of-interest and voxel-based study at 3T and systematic review with meta-analysis. NeuroImage Clin. 3, 481–488 (2013)
Strubell, E., Ganesh, A., McCallum, A.: Energy and policy considerations for deep learning in NLP. In: 57th Meeting of the Association for Computational Linguistics (2019)
Thompson, N.C., Greenewald, K., Lee, K., Manso, G.F.: The computational limits of deep learning. arXiv arXiv:abs/2007.05558 (2022)
Yu, J., et al.: FastFlow: unsupervised anomaly detection and localization via 2D normalizing flows. arXiv arXiv:abs/2111.07677 (2021)
Data Use Declaration and Acknowledgement
G. Oudoumanessah was financially supported by the AURA region. This work has been partially supported by MIAI@Grenoble Alpes (ANR-19-P3IA-0003), and was granted access to the HPC resources of IDRIS under the allocation 2022-AD011013867 made by GENCI. The data used in the preparation of this article were obtained from the Parkinson’s Progression Markers Initiative database www.ppmi-info.org/access-data-specimens/download-data openly available for researchers.
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
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Oudoumanessah, G., Lartizien, C., Dojat, M., Forbes, F. (2023). Towards Frugal Unsupervised Detection of Subtle Abnormalities in Medical Imaging. In: Greenspan, H., et al. Medical Image Computing and Computer Assisted Intervention – MICCAI 2023. MICCAI 2023. Lecture Notes in Computer Science, vol 14222. Springer, Cham. https://doi.org/10.1007/978-3-031-43898-1_40
Download citation
DOI: https://doi.org/10.1007/978-3-031-43898-1_40
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-43897-4
Online ISBN: 978-3-031-43898-1
eBook Packages: Computer ScienceComputer Science (R0)
