GR-Diffusion: 3D Gaussian Representation Meets Diffusion in Whole-Body PET Reconstruction

TL;DR

GR-Diffusion introduces a novel framework combining 3D Gaussian representations with diffusion models to improve low-dose whole-body PET reconstruction, effectively reducing noise and preserving details.

Contribution

This work is the first to integrate 3D Gaussian representations with diffusion models for PET reconstruction, providing a physically grounded and structurally explicit prior.

Findings

Outperforms state-of-the-art methods in image quality

Preserves physiological details effectively

Enhances 3D PET reconstruction at low doses

Abstract

Positron emission tomography (PET) reconstruction is a critical challenge in molecular imaging, often hampered by noise amplification, structural blurring, and detail loss due to sparse sampling and the ill-posed nature of inverse problems. The three-dimensional discrete Gaussian representation (GR), which efficiently encodes 3D scenes using parameterized discrete Gaussian distributions, has shown promise in computer vision. In this work, we pro-pose a novel GR-Diffusion framework that synergistically integrates the geometric priors of GR with the generative power of diffusion models for 3D low-dose whole-body PET reconstruction. GR-Diffusion employs GR to generate a reference 3D PET image from projection data, establishing a physically grounded and structurally explicit benchmark that overcomes the low-pass limitations of conventional point-based or voxel-based methods. This reference image serves as a dual guide during the diffusion process, ensuring both global consistency and local accuracy. Specifically, we employ a hierarchical guidance mechanism based on the GR reference. Fine-grained guidance leverages differences to refine local details, while coarse-grained guidance uses multi-scale difference maps to correct deviations. This strategy allows the diffusion model to sequentially integrate the strong geometric prior from GR and recover sub-voxel information. Experimental results on the UDPET and Clinical datasets with varying dose levels show that GR-Diffusion outperforms state-of-the-art methods in enhancing 3D whole-body PET image quality and preserving physiological details.