Score-Based Generative Models for PET Image Reconstruction

• DOI: 10.48550/arxiv.2308.14190
• Publication Date: 2024-01-23
• Authors: Riccardo Barbano; Alexander Denker; Imraj RD Singh; Željko Kereta; Bangti Jin; Kris Thielemans; Peter Maass; Simon Arridge
• Keywords: FOS: Computer and information sciences; Positron emission tomography; Computer Science - Machine Learning; J.2; Computer Science - Artificial Intelligence; Computer Vision and Pattern Recognition (cs.CV); Image and Video Processing (eess.IV); Computer Science - Computer Vision and Pattern Recognition; I.4.9; J.2; I.2.1; Electrical Engineering and Systems Science - Image and Video Processing; image reconstruction; I.2.1; Machine Learning (cs.LG); 03 medical and health sciences; Artificial Intelligence (cs.AI); 0302 clinical medicine; 15A29, 45Q05; FOS: Electrical engineering, electronic engineering, information engineering; I.4.9; score-based generative models

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 <emph>in-silico</emph> 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.