Direct PET Image Reconstruction Incorporating Deep Image Prior and a Forward Projection Model

Abstract

Convolutional neural networks (CNNs) have recently achieved remarkable performance in positron emission tomography (PET) image reconstruction. In particular, CNN-based direct PET image reconstruction, which directly generates the reconstructed image from the sinogram, has potential applicability to PET image enhancements because it does not require image reconstruction algorithms, which often produce some artifacts. However, these deep learning-based, direct PET image reconstruction algorithms have the disadvantage that they require a large number of high-quality training datasets. In this study, we propose an unsupervised direct PET image reconstruction method that incorporates a deep image prior framework. Our proposed method incorporates a forward projection model with a loss function to achieve unsupervised direct PET image reconstruction from sinograms. To compare our proposed direct reconstruction method with the filtered back projection (FBP) and maximum likelihood expectation maximization (ML-EM) algorithms, we evaluated using Monte Carlo simulation data of brain [${}^{18}$F]FDG PET scans. The results demonstrate that our proposed direct reconstruction quantitatively and qualitatively outperforms the FBP and ML-EM algorithms with respect to peak signal-to-noise ratio and structural similarity index.

This work has been submitted to the IEEE for possible publication

References in corpus (2)

• Anatomical-Guided Attention Enhances Unsupervised PET Image Denoising Performance
• DUG-RECON: A Framework for Direct Image Reconstruction using Convolutional Generative Networks

Cited by in corpus (7)

• List-Mode PET Image Reconstruction Using Deep Image Prior
• Fully 3D Implementation of the End-to-end Deep Image Prior-based PET Image Reconstruction Using Block Iterative Algorithm
• Self-Supervised Pre-Training for Deep Image Prior-Based Robust PET Image Denoising
• ReconU-Net: a direct PET image reconstruction using U-Net architecture with back projection-induced skip connection
• List-Mode PET Image Reconstruction Using Dykstra-Like Splitting
• Limited-angle SPECT image reconstruction using deep image prior
• Exploiting network optimization stability for enhanced PET image denoising using deep image prior