Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2509.14827 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908545781334016 |
|---|---|
| author | Madlindl, Patrick Bongratz, Fabian Wachinger, Christian |
| author_facet | Madlindl, Patrick Bongratz, Fabian Wachinger, Christian |
| contents | Cortical surface reconstruction (CSR) from magnetic resonance imaging (MRI) is fundamental to neuroimage analysis, enabling morphological studies of the cerebral cortex and functional brain mapping. Recent advances in learning-based CSR have dramatically accelerated processing, allowing for reconstructions through the deformation of anatomical templates within seconds. However, ensuring the learned deformations are optimal in terms of deformation energy and consistent across training runs remains a particular challenge. In this work, we design a Minimal Energy Deformation (MED) loss, acting as a regularizer on the deformation trajectories and complementing the widely used Chamfer distance in CSR. We incorporate it into the recent V2C-Flow model and demonstrate considerable improvements in previously neglected training consistency and reproducibility without harming reconstruction accuracy and topological correctness. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_14827 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Template-Based Cortical Surface Reconstruction with Minimal Energy Deformation Madlindl, Patrick Bongratz, Fabian Wachinger, Christian Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning Neurons and Cognition Cortical surface reconstruction (CSR) from magnetic resonance imaging (MRI) is fundamental to neuroimage analysis, enabling morphological studies of the cerebral cortex and functional brain mapping. Recent advances in learning-based CSR have dramatically accelerated processing, allowing for reconstructions through the deformation of anatomical templates within seconds. However, ensuring the learned deformations are optimal in terms of deformation energy and consistent across training runs remains a particular challenge. In this work, we design a Minimal Energy Deformation (MED) loss, acting as a regularizer on the deformation trajectories and complementing the widely used Chamfer distance in CSR. We incorporate it into the recent V2C-Flow model and demonstrate considerable improvements in previously neglected training consistency and reproducibility without harming reconstruction accuracy and topological correctness. |
| title | Template-Based Cortical Surface Reconstruction with Minimal Energy Deformation |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning Neurons and Cognition |
| url | https://arxiv.org/abs/2509.14827 |