Abstract
Generative shape models are crucial for many medical image analysis tasks. In previous studies, it has been shown that conventional methods like PCA-based statistical shape models (SSMs) and their extensions are thought to be robust in terms of generalization ability but have rather poor specificity. On the contrary, deep learning approaches like autoencoders, require large training set sizes, but are comparably specific. In this work, we comprehensively compare different classical and deep learning-based generative shape modeling approaches and demonstrate their limitations and advantages. Experiments on a publicly available 2D chest X-ray data set show that the deep learning methods achieve better specificity and similar generalization abilities for large training set sizes. Furthermore, an extensive analysis of the different methods, gives an insight on their latent space representations.
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
Cootes TF, Taylor CJ, Cooper DH, et al. Active shape models-their training and application. Comput Vis Image Underst. 1995;61(1):38–59.
Kirschner M, Becker M, Wesarg S. 3D active shape model segmentation with nonlinear shape priors. Proc MICCAI. 2011; p. 492–499.
Krüger J, Ehrhardt J, Handels H. Statistical appearance models based on probabilistic correspondences. Med Image Anal. 2017;37:146–159.
Davatzikos C, Tao X, Shen D. Hierarchical active shape models, using the wavelet transform. IEEE Trans Med Imaging; p. 2003.
Wilms M, Handels H, Ehrhardt J. Multi-resolution multi-object statistical shape models based on the locality assumption. Med Image Anal. 2017;38:17–29.
Goodfellow I, Pouget-Abadie J, Mirza M, et al. Generative adversarial nets. In: Advances in Neural Information Processing Systems; 2014. p. 2672–2680.
Ruan X, Murphy RF. Evaluation of methods for generative modeling of cell and nuclear shape. Bioinformatics. 2018 12;35(14):2475–2485.
Uzunova H, Kaftan P, Wilms M, et al. Quantitative comparison of generative shape models for medical images. In: BVM; 2020. p. 201–207.
van Ginneken B, Stegmann MB, Loog M. Segmentation of anatomical structures in chest radiographs using supervised methods. Med Image Anal. 2006; p. 19–40.
Wilms M, Ehrhardt J, Forkert ND. A kernelized multi-level localization method for exible shape modeling with few training data. Proc MICCAI. 2020; p. 765–775.
Kingma D, Welling M. Auto-encoding variational bayes. In: International Conference on Learning Representations; 2014. p. 1–10.
Shu Z, Sahasrabudhe M, Alp Güler R, et al. Deforming autoencoders: unsupervised disentangling of shape and appearance. Proc ECCV. 2018; p. 664–680.
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
Uzunova, H. et al. (2021). Analysis of Generative Shape Modeling Approaches. 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_84
Download citation
DOI: https://doi.org/10.1007/978-3-658-33198-6_84
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)