A Soft-Correspondence Approach to Shape-based Disease Grading with Graph Convolutional Networks
- 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 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.
Author: | Julius Mayer, Daniel BaumORCiD, Felix AmbellanORCiD, Christoph von Tycowicz |
---|---|
Document Type: | In Proceedings |
Parent Title (English): | Proceedings of Machine Learning Research |
Volume: | 194 |
First Page: | 85 |
Last Page: | 95 |
Fulltext Url: | https://openreview.net/pdf?id=Wa6mfpfr29o |
Year of first publication: | 2022 |