Structure-Adaptive Sparse Diffusion in Voxel Space for 3D Medical Image Enhancement

• URL: http://arxiv.org/abs/2604.17773v1
• Date: Mon, 20 Apr 2026 03:52:49 GMT
• Title: Structure-Adaptive Sparse Diffusion in Voxel Space for 3D Medical Image Enhancement
• Abstract summary: Three-dimensional (3D) medical image enhancement, including denoising and super-resolution, is critical for clinical diagnosis in CT, PET, and MRI.<n>We propose a sparse voxel-space diffusion framework that trains and samples on a compact set of uniformly subsampled timesteps.<n>Our framework preserves fine anatomical detail without lossy compression while achieving up to $10times$ training acceleration.
• Score: 10.028882244604809
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Three-dimensional (3D) medical image enhancement, including denoising and super-resolution, is critical for clinical diagnosis in CT, PET, and MRI. Although diffusion models have shown remarkable success in 2D medical imaging, scaling them to high-resolution 3D volumes remains computationally prohibitive due to lengthy diffusion trajectories over high-dimensional volumetric data. We observe that in conditional enhancement, strong anatomical priors in the degraded input render dense noise schedules largely redundant. Leveraging this insight, we propose a sparse voxel-space diffusion framework that trains and samples on a compact set of uniformly subsampled timesteps. The network predicts clean data directly on the data manifold, supervised in velocity space for stable gradient scaling. A lightweight Structure-aware Trajectory Modulation (STM) module recalibrates time embeddings at each network block based on local anatomical content, enabling structure-adaptive denoising over the shared sparse schedule. Operating directly in voxel space, our framework preserves fine anatomical detail without lossy compression while achieving up to $10\times$ training acceleration. Experiments on four datasets spanning CT, PET, and MRI demonstrate state-of-the-art performance on both denoising and super-resolution tasks. Our code is publicly available at: https://github.com/mirthAI/sparse-3d-diffusion.

Related papers

• VoxStruct3D: Structure-Leading Flow Matching for Voxel-Space 3D MRI Synthesis [33.836541706312055]
  High-fidelity 3D MRI synthesis requires both globally coherent anatomy and fine-grained voxel-level detail.<n>We present VoxStruct3D, a voxel-space flow-matching framework that directly models full-resolution MRI volumes.<n> Experiments on pathological and healthy T1-weighted brain MRI datasets show that VoxStruct3D achieves the strongest overall performance.
  arXiv  Detail & Related papers  (2026-08-05T07:51:35Z)
• PPDM: Pixel Puzzling Diffusion Model for Speed and Memory Efficient Volumetric Medical Image Translation [16.556040700298027]
  Pixel Puzzling Diffusion Model (PPDM) is a framework for memory- and speed-efficient 3D medical image translation.<n>We evaluate PPDM on multiple challenging 3D medical image translation tasks, including low-count PET denoising and joint PET denoising and attenuation correction.
  arXiv  Detail & Related papers  (2026-06-13T14:32:03Z)
• WaveDiT: Distribution-Aware Wavelet Flow Matching for Efficient 3D Brain MRI Synthesis [7.371811584771131]
  Large demographically balanced datasets are essential for reliable neuroimaging biomarkers.<n>WaveDiT is a conditional flow matching framework operating in the coefficient space of a 3D Haar Discrete Wavelet Transform.<n>It supports full-resolution 3D synthesis under practical and memory time constraints on a single modern GPU.
  arXiv  Detail & Related papers  (2026-06-07T15:21:02Z)
• EPC-3D-Diff: Equivariant Physics Consistent Conditional 3D Latent Diffusion for CBCT to CT Synthesis [36.790666885042356]
  We propose EPC-3D-Diff, a novel conditional 3D latent diffusion framework for CBCT to CT synthesis.<n>We exploit the fact that an in plane rotation of the volume corresponds to an angular shift in its projections.<n>We validate on a paired head CBCT/CT phantom dataset, including repeat scans, and paired clinical data using patient wise splits.<n> EPC-3D-Diff generalizes well and achieved substantial improvements, +7.4 dB (phantom) and +1.8 dB (clinical data) in PSNR compared to state of the art methods.
  arXiv  Detail & Related papers  (2026-05-19T20:34:09Z)
• Conditional Diffusion Posterior Alignment for Sparse-View CT Reconstruction [4.107377779783183]
  We propose Conditional Posterior Alignment (CDPA) to enable scalable 3D sparse-view CT reconstruction.<n>A 2D U-Net diffusion model is conditioned on an initial 3D reconstruction to improve inter-slice consistency.<n> Experiments on synthetic and real Cone Beam CT (CBCT) data show state-of-the-art performance.
  arXiv  Detail & Related papers  (2026-04-23T14:09:13Z)
• MedPruner: Training-Free Hierarchical Token Pruning for Efficient 3D Medical Image Understanding in Vision-Language Models [59.180043227905294]
  We propose MedPruner, a training-free and model-agnostic hierarchical token pruning framework for efficient 3D medical image understanding.<n>We show that MedPruner enables models such as MedGemma to maintain or even exceed their original performance while retaining fewer than 5% of visual tokens.
  arXiv  Detail & Related papers  (2026-03-12T07:37:00Z)
• Improving 2D Diffusion Models for 3D Medical Imaging with Inter-Slice Consistent Stochasticity [47.52394044948656]
  Inter-Slice Consistentity (ISCS) is a strategy that encourages interslice consistency during diffusion sampling.<n>ISCS can be dropped into any 2D trained diffusion based 3D reconstruction pipeline without additional computational cost.
  arXiv  Detail & Related papers  (2026-02-04T02:54:08Z)
• Accelerating 3D Photoacoustic Computed Tomography with End-to-End Physics-Aware Neural Operators [74.65171736966131]
  Photoacoustic computed tomography (PACT) combines optical contrast with ultrasonic resolution, achieving deep-tissue imaging beyond the optical diffusion limit.<n>Current implementations require dense transducer arrays and prolonged acquisition times, limiting clinical translation.<n>We introduce Pano, an end-to-end physics-aware model that directly learns the inverse acoustic mapping from sensor measurements to volumetric reconstructions.
  arXiv  Detail & Related papers  (2025-09-11T23:12:55Z)
• SUFFICIENT: A scan-specific unsupervised deep learning framework for high-resolution 3D isotropic fetal brain MRI reconstruction [7.268308489093152]
  We propose an unsupervised iterative SVR-SRR framework for isotropic HR volume reconstruction.<n>A decoding network embedded within a deep image prior framework is incorporated with a comprehensive image degradation model to produce the high-resolution (HR) volume.<n>Experiments conducted on large-magnitude motion-corrupted simulation data and clinical data demonstrate the superior performance of the proposed framework.
  arXiv  Detail & Related papers  (2025-05-23T04:53:59Z)
• Introducing 3D Representation for Medical Image Volume-to-Volume Translation via Score Fusion [3.3559609260669303]
  We present Score-Fusion, a novel volumetric translation model that effectively learns 3D representations by ensembling perpendicularly trained 2D diffusion models in score function space.<n>We show that Score-Fusion achieves superior accuracy and volumetric fidelity in 3D medical image super-resolution and modality translation.
  arXiv  Detail & Related papers  (2025-01-13T15:54:21Z)
• StableDreamer: Taming Noisy Score Distillation Sampling for Text-to-3D [88.66678730537777]
  We present StableDreamer, a methodology incorporating three advances. First, we formalize the equivalence of the SDS generative prior and a simple supervised L2 reconstruction loss. Second, our analysis shows that while image-space diffusion contributes to geometric precision, latent-space diffusion is crucial for vivid color rendition.
  arXiv  Detail & Related papers  (2023-12-02T02:27:58Z)
• DEFN: Dual-Encoder Fourier Group Harmonics Network for Three-Dimensional Indistinct-Boundary Object Segmentation [6.0920148653974255]
  We introduce Defect Injection (SDi) to augment the representational diversity of challenging indistinct-boundary objects within training corpora. Consequently, we propose the Dual-Encoder Fourier Group Harmonics Network (DEFN) to tailor incorporating noise, amplify detailed feature recognition, and bolster representation across diverse medical imaging scenarios.
  arXiv  Detail & Related papers  (2023-11-01T12:33:04Z)
• 4D Spatio-Temporal Convolutional Networks for Object Position Estimation in OCT Volumes [69.62333053044712]
  3D convolutional neural networks (CNNs) have shown promising performance for pose estimation of a marker object using single OCT images. We extend 3D CNNs to 4D-temporal CNNs to evaluate the impact of additional temporal information for marker object tracking.
  arXiv  Detail & Related papers  (2020-07-02T12:02:20Z)
• Multifold Acceleration of Diffusion MRI via Slice-Interleaved Diffusion Encoding (SIDE) [50.65891535040752]
  We propose a diffusion encoding scheme, called Slice-Interleaved Diffusion. SIDE, that interleaves each diffusion-weighted (DW) image volume with slices encoded with different diffusion gradients. We also present a method based on deep learning for effective reconstruction of DW images from the highly slice-undersampled data.
  arXiv  Detail & Related papers  (2020-02-25T14:48:17Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.