A Soft-Correspondence Approach to Shape-based Disease Grading with Graph Convolutional Networks

Julius Mayer, Daniel Baum, Felix Ambellan, Christoph von von Tycowicz
Proceedings of the First International Workshop on Geometric Deep Learning in Medical Image Analysis, PMLR 194:85-95, 2022.

Abstract

Shape analysis provides principled means for understanding anatomical structures from medical images. The underlying notions of shape spaces, however, come with strict assumptions prohibiting the analysis of incomplete and/or topologically varying shapes. This work aims to alleviate these limitations by adapting the concept of \textit{soft correspondences}. In particular, we present a graph-based learning approach for morphometric classification of disease states that is based on a generalized notion of shape correspondences in terms of functional maps. We demonstrate the performance of the derived classifier on the open-access ADNI database for differentiating normal controls and subjects with Alzheimer’s disease. Notably, our experiment shows that our approach can improve over state-of-the-art from geometric deep learning.

Cite this Paper


BibTeX
@InProceedings{pmlr-v194-mayer22a, title = {A Soft-Correspondence Approach to Shape-based Disease Grading with Graph Convolutional Networks}, author = {Mayer, Julius and Baum, Daniel and Ambellan, Felix and von Tycowicz, Christoph von}, booktitle = {Proceedings of the First International Workshop on Geometric Deep Learning in Medical Image Analysis}, pages = {85--95}, year = {2022}, editor = {Bekkers, Erik and Wolterink, Jelmer M. and Aviles-Rivero, Angelica}, volume = {194}, series = {Proceedings of Machine Learning Research}, month = {18 Nov}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v194/mayer22a/mayer22a.pdf}, url = {https://proceedings.mlr.press/v194/mayer22a.html}, abstract = {Shape analysis provides principled means for understanding anatomical structures from medical images. The underlying notions of shape spaces, however, come with strict assumptions prohibiting the analysis of incomplete and/or topologically varying shapes. This work aims to alleviate these limitations by adapting the concept of \textit{soft correspondences}. In particular, we present a graph-based learning approach for morphometric classification of disease states that is based on a generalized notion of shape correspondences in terms of functional maps. We demonstrate the performance of the derived classifier on the open-access ADNI database for differentiating normal controls and subjects with Alzheimer’s disease. Notably, our experiment shows that our approach can improve over state-of-the-art from geometric deep learning. } }
Endnote
%0 Conference Paper %T A Soft-Correspondence Approach to Shape-based Disease Grading with Graph Convolutional Networks %A Julius Mayer %A Daniel Baum %A Felix Ambellan %A Christoph von von Tycowicz %B Proceedings of the First International Workshop on Geometric Deep Learning in Medical Image Analysis %C Proceedings of Machine Learning Research %D 2022 %E Erik Bekkers %E Jelmer M. Wolterink %E Angelica Aviles-Rivero %F pmlr-v194-mayer22a %I PMLR %P 85--95 %U https://proceedings.mlr.press/v194/mayer22a.html %V 194 %X Shape analysis provides principled means for understanding anatomical structures from medical images. The underlying notions of shape spaces, however, come with strict assumptions prohibiting the analysis of incomplete and/or topologically varying shapes. This work aims to alleviate these limitations by adapting the concept of \textit{soft correspondences}. In particular, we present a graph-based learning approach for morphometric classification of disease states that is based on a generalized notion of shape correspondences in terms of functional maps. We demonstrate the performance of the derived classifier on the open-access ADNI database for differentiating normal controls and subjects with Alzheimer’s disease. Notably, our experiment shows that our approach can improve over state-of-the-art from geometric deep learning.
APA
Mayer, J., Baum, D., Ambellan, F. & von Tycowicz, C.v.. (2022). A Soft-Correspondence Approach to Shape-based Disease Grading with Graph Convolutional Networks. Proceedings of the First International Workshop on Geometric Deep Learning in Medical Image Analysis, in Proceedings of Machine Learning Research 194:85-95 Available from https://proceedings.mlr.press/v194/mayer22a.html.

Related Material