Score-Based Generative Models for PET Image Reconstruction

Abstract

Score-based generative models have demonstrated highly promising results for medical image reconstruction tasks in magnetic resonance imaging or computed tomography. However, their application to Positron Emission Tomography (PET) is still largely unexplored. PET image reconstruction involves a variety of challenges, including Poisson noise with high variance and a wide dynamic range. To address these challenges, we propose several PET-specific adaptations of score-based generative models. The proposed framework is developed for both 2D and 3D PET. In addition, we provide an extension to guided reconstruction using magnetic resonance images. We validate the approach through extensive 2D and 3D $\text{in-silico}$ experiments with a model trained on patient-realistic data without lesions, and evaluate on data without lesions as well as out-of-distribution data with lesions. This demonstrates the proposed method's robustness and significant potential for improved PET reconstruction.

arXiv commentAccepted for publication at the Journal of Machine Learning for Biomedical Imaging (MELBA) https://melba-journal.org/2024:001

References in corpus (29)

• Auto-Encoding Variational Bayes
• Denoising Diffusion Probabilistic Models
• Deep Convolutional Neural Network for Inverse Problems in Imaging
• Diffusion Models Beat GANs on Image Synthesis
• Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding
• Deep Unsupervised Learning using Nonequilibrium Thermodynamics
• Score-Based Generative Modeling through Stochastic Differential Equations
• Generative Modeling by Estimating Gradients of the Data Distribution
• On instabilities of deep learning in image reconstruction - Does AI come at a cost?
• Learned Primal-dual Reconstruction
• Classifier-Free Diffusion Guidance
• Diffusion-LM Improves Controllable Text Generation
• Diffusion Posterior Sampling for General Noisy Inverse Problems
• Tackling the Generative Learning Trilemma with Denoising Diffusion GANs
• Solving 3D Inverse Problems using Pre-trained 2D Diffusion Models
• Solving Inverse Problems in Medical Imaging with Score-Based Generative Models
• Denoising Diffusion Implicit Models
• Maximum Likelihood Training of Score-Based Diffusion Models
• Robust Compressed Sensing MRI with Deep Generative Priors
• List-Mode PET Image Reconstruction Using Deep Image Prior
• PARALLELPROJ -- An open-source framework for fast calculation of projections in tomography
• Brain Imaging Generation with Latent Diffusion Models
• PET image denoising based on denoising diffusion probabilistic models
• Denoising Diffusion Models for Plug-and-Play Image Restoration
• Test-Time Training Can Close the Natural Distribution Shift Performance Gap in Deep Learning Based Compressed Sensing
• Optimizing Sampling Patterns for Compressed Sensing MRI with Diffusion Generative Models
• Reflected Diffusion Models
• MRI Reconstruction with Side Information using Diffusion Models
• Learning Gradually Non-convex Image Priors Using Score Matching

Cited by in corpus (6)

• A Review on Low-Dose Emission Tomography Post-Reconstruction Denoising with Neural Network Approaches
• Artificial Intelligence-Guided PET Image Reconstruction and Multi-Tracer Imaging: Novel Methods, Challenges, And Opportunities
• Generative-Model-Based Fully 3D PET Image Reconstruction by Conditional Diffusion Sampling
• Multi-Subject Image Synthesis as a Generative Prior for Single-Subject PET Image Reconstruction
• Supervised Diffusion-Model-Based PET Image Reconstruction
• Bayesian PET Reconstruction with Learned Flow Matching Priors via Langevin Sampling