Research map / Image generation / Fast diffusion sampling
A Diffusion Model for Simulation Ready Coronary Anatomy with Morpho-skeletal Control
Karim Kadry, Shreya Gupta, Jonas Sogbadji, Michiel Schaap, Kersten Petersen, Takuya Mizukami, Carlos Collet, Farhad R. Nezami, Elazer R Edelman
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add topological interaction loss to VAE training (ablation) | VAE reconstruction · 100 training maps | topological violations, lumen (%): 17 -> 0.5 | Improved |
| train LDM on topologically regularized VAE latents (ablation) | unconditional LDM · 200 sampled arteries | topological violations, calcium (%): 9.21 -> 2.61 | Improved |
| add morphological conditioning to unconditional LDM (ablation) | LDM · validation conditions | morphological fidelity (MAE): 117 -> 4.0 | Improved |
| add morphological conditioning alone, effect on skeletal fidelity (ablation) | LDM · validation conditions | skeletal fidelity (branch count MAE): 3.2 -> 1.2 | Improved |
| joint morpho-skeletal conditioning vs morphology alone (ablation) | LDM · validation conditions | skeletal fidelity and 2D recall: skeletal 1.2 -> 0.48; 2D recall 0.82 -> 0.86 | Improved |
| increase classifier-free guidance weight with unconditional model (ablation) | CFG · morpho-skeletal guidance | sample quality | Hurt |
| use adaptive null guidance instead of loss guidance (CG-Soft, DPS) | morpho-skeletal LDM · sampling cost | sample time (s) / memory (GB): 89 -> 46 s; 15.1 -> 6.4 GB vs DPS-Soft | Improved |
| increase ANG guidance weight (ablation) | ANG · morpho-skeletal guidance | topological violations and 3D morphological precision | Hurt |
| use DPS-Soft differentiable regressor guidance | morpho-skeletal LDM · qualitative | visual quality | Mixed |
Closest papers by meaning
- VasTSD: Learning 3D Vascular Tree-state Space Diffusion Model for Angiography Synthesis (CVPR 2025 · similarity 0.84)
- Variational Diffusion Posterior Sampling with Midpoint Guidance (ICLR 2025 · similarity 0.80)
- AnaDiffusion: Anatomically Compositional Latent Diffusion for Controllable 3D Brain MRI Generation (NeurIPS 2026 · similarity 0.80)
- Training-Free Constrained Generation With Stable Diffusion Models (NeurIPS 2025 · similarity 0.80)
- AMD: Anatomical Motion Diffusion with Interpretable Motion Decomposition and Fusion (AAAI 2024 · similarity 0.80)
- SurgVista: Long-Horizon Surgical World Modeling with Plausible Instrument-Tissue Dynamics (NeurIPS 2026 · similarity 0.80)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
