Feed-Forward Hierarchical Gaussian Diffusion for Extreme CT Reconstruction

• URL: http://arxiv.org/abs/2608.15343v1
• Date: Sat, 15 Aug 2026 17:56:09 GMT
• Title: Feed-Forward Hierarchical Gaussian Diffusion for Extreme CT Reconstruction
• Abstract summary: HiGDiff is a feed-forward hierarchical Gaussian diffusion framework that decomposes reconstruction both spatially and from structure to detail.<n> Experiments on three distinct CT benchmark datasets demonstrate state-of-the-art reconstruction performance across isolated, paired, and joint degradation settings.
• Score: 5.827215998326956
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Reconstructing three-dimensional computed tomography (CT) from severely constrained projections is highly ill-posed. Sparse angular sampling, restricted angular coverage, and low photon counts can occur individually or jointly, obscuring global anatomy and local tissue detail. Many learned CT reconstruction methods are tailored to a single dominant degradation. Existing diffusion and Gaussian approaches commonly recover global structure and local detail within a shared representation. We propose HiGDiff, a feed-forward hierarchical Gaussian diffusion framework that decomposes reconstruction both spatially and from structure to detail. Physics-conditioned anatomical anchors and a foreground capacity field allocate learnable Gaussian primitives to informative regions. A structure diffusion stage first recovers global attenuation geometry, and its learned representation conditions a detail diffusion stage for residual boundaries and tissue transitions. The resulting Gaussian banks are rendered as attenuation fields and further refined by a gradient-isolated residual module. Experiments on three distinct CT benchmark datasets demonstrate state-of-the-art reconstruction performance across isolated, paired, and joint degradation settings, including improvements of 5.81 dB in macro-average peak signal-to-noise ratio (PSNR) and 0.113 in structural similarity index measure (SSIM) on the Low Dose CT Image and Projection Data (LDCT-PD) collection. Code and experimental configurations are openly available at https://github.com/Bean-Young/HiGDiff.

Related papers

• Reconstruction-Shift Discrimination via Mask-Guided Latent Diffusion for Medical Anomaly Detection [58.92871907037662]
  Unsupervised medical anomaly detection learns normal anatomical patterns from healthy training images and identifies deviations at test time.<n>Reconstruction-based and diffusion-based methods commonly use the difference between an input image and its reconstruction as anomaly evidence.<n>We propose discriminative mask-guided diffusion (DMD), a medical anomaly detection framework that complements residual-based localization with reconstruction-shift discrimination.
  arXiv  Detail & Related papers  (2026-08-01T04:56:39Z)
• GenDiff: A Dose and Anatomy Aware Diffusion Model with Structural Prior Refinement for Low-Dose CT Reconstruction and Generalization [0.0]
  We propose GenDiff, a generalizable diffusion-based framework for low-dose CT reconstruction.<n>We show that GenDiff consistently outperforms state-of-the-art convolutional neural network and diffusion-based reconstruction methods.<n>The proposed approach achieves superior reconstruction quality while maintaining strong robustness across different dose levels, anatomical regions, and acquisition domains.
  arXiv  Detail & Related papers  (2026-07-11T09:05:51Z)
• PointDiT: Pixel-Space Diffusion for Monocular Geometry Estimation [105.86385617182879]
  State-of-the-art single-image 3D reconstruction methods often rely on complex hybrid architectures and loss functions.<n>We show that such architectural overhead and intricate loss formulations are unnecessary.<n>We introduce a minimalist pixel-space Diffusion Transformer, built on a plain ViT, that operates directly on raw 3D point map patches.
  arXiv  Detail & Related papers  (2026-07-02T17:59:56Z)
• FSP-Diff: Full-Spectrum Prior-Enhanced DualDomain Latent Diffusion for Ultra-Low-Dose Spectral CT Reconstruction [20.226435074884808]
  FSP-Diff is a full-spectrum prior-enhanced dual-domain latent diffusion framework for ultra-low-dose spectral CT reconstruction.<n>Our framework integrates direct image reconstructions with projection-domain denoised results.<n>Experiments on simulated and real-world datasets demonstrate that FSP-Diff significantly outperforms state-of-the-art methods in both image quality and computational efficiency.
  arXiv  Detail & Related papers  (2026-02-08T14:13:19Z)
• Advanced Geometric Correction Algorithms for 3D Medical Reconstruction: Comparison of Computed Tomography and Macroscopic Imaging [0.9395222766576343]
  This paper introduces a hybrid two-stage registration framework for reconstructing 3D kidney anatomy from macroscopic slices.<n>It addresses the data-scarcity and high-distortion challenges typical of macroscopic imaging.<n>The proposed framework generalizes to other soft-tissue organs reconstructed from optical or photographic cross-sections.
  arXiv  Detail & Related papers  (2026-01-30T17:16:17Z)
• Resolution-Independent Neural Operators for Multi-Rate Sparse-View CT [67.14700058302016]
  Deep learning methods achieve high-fidelity reconstructions but often overfit to a fixed acquisition setup.<n>We propose Computed Tomography neural Operator (CTO), a unified CT reconstruction framework that extends to continuous function space.<n>CTO enables consistent multi-sampling-rate and cross-resolution performance, with on average >4dB PSNR gain over CNNs.
  arXiv  Detail & Related papers  (2025-12-13T08:31:46Z)
• Moving Beyond Diffusion: Hierarchy-to-Hierarchy Autoregression for fMRI-to-Image Reconstruction [65.67001243986981]
  We propose MindHier, a coarse-to-fine fMRI-to-image reconstruction framework built on scale-wise autoregressive modeling.<n>MindHier achieves superior semantic fidelity, 4.67x faster inference, and more deterministic results than the diffusion-based baselines.
  arXiv  Detail & Related papers  (2025-10-25T15:40:07Z)
• CoCPF: Coordinate-based Continuous Projection Field for Ill-Posed Inverse Problem in Imaging [78.734927709231]
  Sparse-view computed tomography (SVCT) reconstruction aims to acquire CT images based on sparsely-sampled measurements. Due to ill-posedness, implicit neural representation (INR) techniques may leave considerable holes'' (i.e., unmodeled spaces) in their fields, leading to sub-optimal results. We propose the Coordinate-based Continuous Projection Field (CoCPF), which aims to build hole-free representation fields for SVCT reconstruction.
  arXiv  Detail & Related papers  (2024-06-21T08:38:30Z)
• MSDiff: Multi-Scale Diffusion Model for Ultra-Sparse View CT Reconstruction [5.5805994093893885]
  We propose an ultra-sparse view CT reconstruction method utilizing multi-scale dif-fusion models (MSDiff) The proposed model ingeniously integrates information from both comprehensive sampling and selectively sparse sampling tech-niques. By leveraging the inherent correlations within the projec-tion data, we have designed an equidistant mask, enabling the model to focus its attention more effectively.
  arXiv  Detail & Related papers  (2024-05-09T14:52:32Z)
• DPER: Diffusion Prior Driven Neural Representation for Limited Angle and Sparse View CT Reconstruction [45.00528216648563]
  Diffusion Prior Driven Neural Representation (DPER) is an unsupervised framework designed to address the exceptionally ill-posed CT reconstruction inverse problems. DPER adopts the Half Quadratic Splitting (HQS) algorithm to decompose the inverse problem into data fidelity and distribution prior sub-problems. We conduct comprehensive experiments to evaluate the performance of DPER on LACT and ultra-SVCT reconstruction with two public datasets.
  arXiv  Detail & Related papers  (2024-04-27T12:55:13Z)
• DiAD: A Diffusion-based Framework for Multi-class Anomaly Detection [55.48770333927732]
  We propose a Difusion-based Anomaly Detection (DiAD) framework for multi-class anomaly detection. It consists of a pixel-space autoencoder, a latent-space Semantic-Guided (SG) network with a connection to the stable diffusion's denoising network, and a feature-space pre-trained feature extractor. Experiments on MVTec-AD and VisA datasets demonstrate the effectiveness of our approach.
  arXiv  Detail & Related papers  (2023-12-11T18:38:28Z)

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.