Physics-Constrained Diffusion Reconstruction with Posterior Correction for Quantitative and Fast PET Imaging

2025-08-19

AI Summary

Deep learning-based PET reconstruction suffers from quantitative inaccuracy, prominent artifacts, and poor interpretability. Method: We propose a physics-constrained conditional diffusion model that embeds a posterior physical correction mechanism within the diffusion sampling process. The model jointly accounts for scatter, attenuation, and random coincidences, and incorporates time-of-flight (TOF) geometric probability images as conditional inputs, augmented by a GTP-image normalization strategy to enhance physical consistency. Results: Our method achieves quantitative accuracy comparable to fully corrected OSEM on both brain and whole-body PET datasets—outperforming end-to-end deep learning models and, in key metrics, surpassing conventional iterative methods. Reconstruction speed improves by 50–85%. Phantom studies demonstrate superior background uniformity and lesion contrast. Moreover, the approach significantly enhances out-of-distribution generalization and clinical interpretability.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Calibration & Uncertainty Quantification
• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Security and Privacy: Data transparency and provenance
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Deep learning-based reconstruction of positron emission tomography(PET) data has gained increasing attention in recent years. While these methods achieve fast reconstruction,concerns remain regarding quantitative accuracy and the presence of artifacts,stemming from limited model interpretability,data driven dependence, and overfitting risks.These challenges have hindered clinical adoption.To address them,we propose a conditional diffusion model with posterior physical correction (PET-DPC) for PET image reconstruction. An innovative normalization procedure generates the input Geometric TOF Probabilistic Image (GTP-image),while physical information is incorporated during the diffusion sampling process to perform posterior scatter,attenuation,and random corrections. The model was trained and validated on 300 brain and 50 whole-body PET datasets,a physical phantom,and 20 simulated brain datasets. PET-DPC produced reconstructions closely aligned with fully corrected OSEM images,outperforming end-to-end deep learning models in quantitative metrics and,in some cases, surpassing traditional iterative methods. The model also generalized well to out-of-distribution(OOD) data. Compared to iterative methods,PET-DPC reduced reconstruction time by 50% for brain scans and 85% for whole-body scans. Ablation studies confirmed the critical role of posterior correction in implementing scatter and attenuation corrections,enhancing reconstruction accuracy. Experiments with physical phantoms further demonstrated PET-DPC's ability to preserve background uniformity and accurately reproduce tumor-to-background intensity ratios. Overall,these results highlight PET-DPC as a promising approach for rapid, quantitatively accurate PET reconstruction,with strong potential to improve clinical imaging workflows.

Problem

• Improving quantitative accuracy in PET image reconstruction
• Reducing artifacts from deep learning-based PET reconstruction
• Accelerating reconstruction time while maintaining image quality

Innovation

• Conditional diffusion model with posterior physical correction
• Generates GTP-image input with innovative normalization
• Incorporates physical information during diffusion sampling process