Latent accelerated diffusion-based deformation estimation for real-time volumetric imaging

Objective.The proposed method aims to address the technical limitations in 3D respiratory estimation and image reconstruction from ultra-sparse views, overcoming data-acquisition constraints that often hinder conventional deformable image registration and volumetric imaging in real-time clinical applications.Approach.We propose aLatentAcceleratedDiffusion framework forDeformationEstimation enablingReal-time volumetric imaging (LADDER) framework, which integrates: (1) a deformation network (VoxelMorph) that generates a patient-specific baseline deformation vector field (DVF) from pre-treatment imaging (e.g. end-inhale to end-exhale), and (2) a latent diffusion model (LDM)-based deformable image registration (DIR) model that estimates DVF scaling factors and residual maps to generate intra-treatment real-time DVF and dynamic volumetric images. The LDM compresses the baseline DVF into a compact latent manifold, enabling fast, projection-conditioned refinement guided by anatomical cues. A physics-informed loss enforces anatomical regularity and consistency with measured projections. LADDER was trained on the Learn2Reg dataset and evaluated on 10 DIR-Lab lung datasets across compression factors, spatial down-sampling, projection configurations (single- vs dual-view), and baseline DVF spans.Main results.With a dual-projection input, an end-inhale to end-exhale baseline DVF and a compression and down-sampling factor of 8, on 10 test cases, LADDER achieves a mean target registration error (TRE) of 0.87 ± 0.33 mm, and high volumetric structure similarity (3D SSIM > 0.95) and low volumetric reconstruction error (3D NMSE < 0.006), while maintaining real-time inference (∼0.11-0.12 s). Further analysis shows the range of DVF span impacts deformation accuracy, and the dual-projection input improves deformation fidelity and reduced variability across breathing phases compared to the single-projection.Significance.LADDER enables real-time 3D motion estimation and volumetric reconstruction from ultra-sparse x-ray views by conditioning diffusion in a patient-specific DVF latent space. Its submillimeter TRE and real-time speed indicate strong potential for next-generation motion management and image guidance, supporting on-the-fly anatomical modeling to enhance the safety and efficacy of lung stereotactic body radiation therapy.