Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2403.11790 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912473809944576 |
|---|---|
| author | Pepe, Antonio Schussnig, Richard Li, Jianning Gsaxner, Christina Schmalstieg, Dieter Egger, Jan |
| author_facet | Pepe, Antonio Schussnig, Richard Li, Jianning Gsaxner, Christina Schmalstieg, Dieter Egger, Jan |
| contents | Shape reconstruction from imaging volumes is a recurring need in medical image analysis. Common workflows start with a segmentation step, followed by careful post-processing and,finally, ad hoc meshing algorithms. As this sequence can be timeconsuming, neural networks are trained to reconstruct shapes through template deformation. These networks deliver state-ofthe-art results without manual intervention, but, so far, they have primarily been evaluated on anatomical shapes with little topological variety between individuals. In contrast, other works favor learning implicit shape models, which have multiple benefits for meshing and visualization. Our work follows this direction by introducing deep medial voxels, a semi-implicit representation that faithfully approximates the topological skeleton from imaging volumes and eventually leads to shape reconstruction via convolution surfaces. Our reconstruction technique shows potential for both visualization and computer simulations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_11790 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Deep Medial Voxels: Learned Medial Axis Approximations for Anatomical Shape Modeling Pepe, Antonio Schussnig, Richard Li, Jianning Gsaxner, Christina Schmalstieg, Dieter Egger, Jan Computer Vision and Pattern Recognition Artificial Intelligence Shape reconstruction from imaging volumes is a recurring need in medical image analysis. Common workflows start with a segmentation step, followed by careful post-processing and,finally, ad hoc meshing algorithms. As this sequence can be timeconsuming, neural networks are trained to reconstruct shapes through template deformation. These networks deliver state-ofthe-art results without manual intervention, but, so far, they have primarily been evaluated on anatomical shapes with little topological variety between individuals. In contrast, other works favor learning implicit shape models, which have multiple benefits for meshing and visualization. Our work follows this direction by introducing deep medial voxels, a semi-implicit representation that faithfully approximates the topological skeleton from imaging volumes and eventually leads to shape reconstruction via convolution surfaces. Our reconstruction technique shows potential for both visualization and computer simulations. |
| title | Deep Medial Voxels: Learned Medial Axis Approximations for Anatomical Shape Modeling |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2403.11790 |