Anatomy-Guided Residual Motion Diffusion for Controllable 4D Cardiac MRI Synthesis

2026-06-25

AI Summary

This work addresses the challenges of scarce annotations, cross-scanner domain shift, and privacy constraints in 4D medical imaging by proposing an anatomy-guided cascaded latent diffusion model. The method decouples static anatomical structures from dynamic residual motion in the latent space and leverages clinical priors to synthesize spatiotemporally coherent and anatomically plausible 4D cardiac MRI sequences. The framework integrates a semi-supervised variational autoencoder, a cascaded latent diffusion model (LDM), and an nnU-Net segmentation module. Experimental results demonstrate that the synthesized data significantly enhances cross-vendor generalization: left ventricular Dice score improves by 2.8% and boundary error decreases by 5.4 mm; overall mean Dice increases by 1.4% and Hausdorff distance reduces by 3.0 mm.

Abstract

Developing robust artificial intelligence models for 4D (3D + time) medical imaging is constrained by limited annotated data, inter-device domain shifts, and privacy restrictions. To address this, we propose a 4D controllable generative framework for anatomically consistent data augmentation. A semi-supervised variational autoencoder learns a compact latent representation of anatomical volumes while jointly predicting aligned segmentation masks in a unified framework. Anatomical structure is then disentangled from temporal dynamics through a cascaded latent diffusion model (LDM). A static LDM generates subject-specific anatomy conditioned on clinical priors (diagnosis and volumes measures) and a subsequent motion LDM estimates residual latent motions, ensuring strict temporal coherence across the 4D sequence. The proposed approach was evaluated on cine cardiac MRI as a representative 4D imaging application. Experiments across multiple datasets demonstrate high controllability of static anatomy (Pearson r > 0.8) and strong temporal coherence (FVD = 288.08). In cross-vendor generalization experiments, augmenting training sets with synthetic 4D sequences significantly improves downstream segmentation performance. Using nnU-Net, the proposed augmentation strategy improves the average Dice score by 1.4% and reduces the Hausdorff Distance by 3.0mm compared to training on real data alone, for the left ventricle, Dice improves by 2.8% with a 5.4mm reduction in boundary error. Overall, this framework provides a scalable and controllable solution for 4D medical image synthesis, supporting the development of more robust models with limited annotations and cross-vendor variability. Code available on https://github.com/cyiheng/4DCardiacMRISynthesis.

Problem

• 4D medical imaging
• limited annotated data
• inter-device domain shifts
• privacy restrictions
• data augmentation

Innovation

• anatomy-guided diffusion
• residual motion modeling
• 4D medical image synthesis
• latent disentanglement
• cross-vendor generalization