Analysis of Generative Shape Modeling Approaches: Latent Space Properties and Interpretability
Hristina Uzunova, Jesse Kruse, Paul Kaftan, Matthias Wilms, Nils D. Forkert, Heinz Handels, Jan Ehrhardt
Universität zu Lübeck, Institut für Medizinische Informatik
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 robust in terms of generalization ability for small training set sizes and have rather poor specificity. In contrary, deep learning approaches like autoencoders, require large training set sizes, but are very specific. Based on that, 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.
The author has not agreed to a public release of their video.