Generative-Model-Based Fully 3D PET Image Reconstruction by Conditional Diffusion Sampling

TL;DR

This paper introduces a novel 3D PET image reconstruction method using score-based generative models, demonstrating improved low-dose image quality and uncertainty quantification on real data compared to traditional methods.

Contribution

The authors develop and implement the first SGM-based 3D PET reconstruction method for real data, including low-count scenarios, and analyze its bias, variance, and uncertainty characteristics.

Findings

SGM-based reconstructions outperform traditional methods in low-dose scenarios.

Reconstructed images exhibit lower variance than baseline methods.

Uncertainty images effectively quantify reconstruction confidence.

Abstract

Score-based generative models (SGMs) have recently shown promising results for image reconstruction on simulated positron emission tomography (PET) datasets. In this work we have developed and implemented practical methodology for 3D image reconstruction with SGMs, and perform (to our knowledge) the first SGM-based reconstruction of real fully 3D PET data. We train an SGM on full-count reference brain images, and extend methodology to allow SGM-based reconstructions at very low counts (1% of original, to simulate low-dose or short-duration scanning). We then perform reconstructions for multiple independent realisations of 1% count data, allowing us to analyse the bias and variance characteristics of the method. We sample from the learned posterior distribution of the generative algorithm to calculate uncertainty images for our reconstructions. We evaluate the method's performance on real full- and low-count PET data and compare with conventional OSEM and MAP-EM baselines, showing that our SGM-based low-count reconstructions match full-dose reconstructions more closely and in a bias-variance trade-off comparison, our SGM-reconstructed images have lower variance than existing baselines. Future work will compare to supervised deep-learned methods, with other avenues for investigation including how data conditioning affects the SGM's posterior distribution and the algorithm's performance with different tracers.