CardioComposer: Leveraging Differentiable Geometry for Compositional Control of Anatomical Diffusion Models
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917348121772032 |
|---|---|
| author | Kadry, Karim Goraya, Shoaib Manicka, Ajay Abdelwahed, Abdalla Chutisilp, Naravich Nezami, Farhad Edelman, Elazer |
| author_facet | Kadry, Karim Goraya, Shoaib Manicka, Ajay Abdelwahed, Abdalla Chutisilp, Naravich Nezami, Farhad Edelman, Elazer |
| contents | Generative models of 3D cardiovascular anatomy can synthesize informative structures for clinical research and medical device evaluation, but face a trade-off between geometric controllability and realism. We propose CardioComposer: a programmable, inference-time framework for generating multi-class anatomical label maps from interpretable ellipsoidal primitives. These primitives represent geometric attributes such as the size, shape, and position of discrete substructures. We specifically develop differentiable measurement functions based on voxel-wise geometric moments, enabling loss-based gradient guidance during diffusion model sampling. We demonstrate that these losses can constrain individual geometric attributes in a disentangled manner and provide compositional control over multiple substructures. Finally, we show that our method is compatible with a broad range of anatomical systems containing non-convex substructures, spanning cardiac, vascular, and skeletal organs. We release our code at https://github.com/kkadry/CardioComposer. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_08015 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | CardioComposer: Leveraging Differentiable Geometry for Compositional Control of Anatomical Diffusion Models Kadry, Karim Goraya, Shoaib Manicka, Ajay Abdelwahed, Abdalla Chutisilp, Naravich Nezami, Farhad Edelman, Elazer Image and Video Processing Artificial Intelligence Computer Vision and Pattern Recognition Machine Learning Generative models of 3D cardiovascular anatomy can synthesize informative structures for clinical research and medical device evaluation, but face a trade-off between geometric controllability and realism. We propose CardioComposer: a programmable, inference-time framework for generating multi-class anatomical label maps from interpretable ellipsoidal primitives. These primitives represent geometric attributes such as the size, shape, and position of discrete substructures. We specifically develop differentiable measurement functions based on voxel-wise geometric moments, enabling loss-based gradient guidance during diffusion model sampling. We demonstrate that these losses can constrain individual geometric attributes in a disentangled manner and provide compositional control over multiple substructures. Finally, we show that our method is compatible with a broad range of anatomical systems containing non-convex substructures, spanning cardiac, vascular, and skeletal organs. We release our code at https://github.com/kkadry/CardioComposer. |
| title | CardioComposer: Leveraging Differentiable Geometry for Compositional Control of Anatomical Diffusion Models |
| topic | Image and Video Processing Artificial Intelligence Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2509.08015 |