# Iterative PET Image Reconstruction Using Convolutional Neural Network Representation

arXiv: https://arxiv.org/abs/1710.03344 · HTML: https://mlsanity.com/paper/1710.03344/

## Key facts

- Cited by 21 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 9 Oct 2017.

## Built on by (most cited first, 21 of 21)

- [Spach Transformer: Spatial and Channel-wise Transformer Based on Local and Global Self-attentions for PET Image Denoising](https://mlsanity.com/paper/2209.03300/) (2022)
- [Direct PET Image Reconstruction Incorporating Deep Image Prior and a Forward Projection Model](https://mlsanity.com/paper/2109.00768/) (2021)
- [Direct Reconstruction of Linear Parametric Images from Dynamic PET Using Nonlocal Deep Image Prior](https://mlsanity.com/paper/2106.10359/) (2021)
- [List-Mode PET Image Reconstruction Using Deep Image Prior](https://mlsanity.com/paper/2204.13404/) (2022)
- [Deep Kernel Representation for Image Reconstruction in PET](https://mlsanity.com/paper/2110.01174/) (2021)
- [Fully 3D Implementation of the End-to-end Deep Image Prior-based PET Image Reconstruction Using Block Iterative Algorithm](https://mlsanity.com/paper/2212.11844/) (2022)
- [PET Image Reconstruction Using Deep Diffusion Image Prior](https://mlsanity.com/paper/2507.15078/) (2025)
- [Neural KEM: A Kernel Method with Deep Coefficient Prior for PET Image Reconstruction](https://mlsanity.com/paper/2201.01443/) (2022)
- [Diffusion Transformer Model With Compact Prior for Low-dose PET Reconstruction](https://mlsanity.com/paper/2407.00944/) (2024)
- [Diffusion Transformer Meets Random Masks: An Advanced PET Reconstruction Framework](https://mlsanity.com/paper/2503.08339/) (2025)
- [Unsupervised PET Reconstruction from a Bayesian Perspective](https://mlsanity.com/paper/2110.15568/) (2021)
- [DULDA: Dual-domain Unsupervised Learned Descent Algorithm for PET image reconstruction](https://mlsanity.com/paper/2303.04661/) (2023)
- [Accurate 2D Reconstruction for PET Scanners based on the Analytical White Image Model](https://mlsanity.com/paper/2306.17652/) (2023)
- [Application of Spherical Convolutional Neural Networks to Image Reconstruction and Denoising in Nuclear Medicine](https://mlsanity.com/paper/2307.03298/) (2023)
- [Convergent ADMM Plug and Play PET Image Reconstruction](https://mlsanity.com/paper/2310.04299/) (2023)
- [ReconU-Net: a direct PET image reconstruction using U-Net architecture with back projection-induced skip connection](https://mlsanity.com/paper/2312.02494/) (2023)
- [Deep kernel representations of latent space features for low-dose PET-MR imaging robust to variable dose reduction](https://mlsanity.com/paper/2409.06198/) (2024)
- [Dynamic PET Image Reconstruction via Non-negative INR Factorization](https://mlsanity.com/paper/2503.08025/) (2025)
- [Double-Constraint Diffusion Model with Nuclear Regularization for Ultra-low-dose PET Reconstruction](https://mlsanity.com/paper/2509.00395/) (2025)
- [Generalizable CT-Free PET Attenuation and Scatter Correction for Pediatric Patients](https://mlsanity.com/paper/2604.22894/) (2026)
- [Flow Matching-Based PET Image Reconstruction](https://mlsanity.com/paper/2608.20112/) (2026)

## Abstract (arXiv)

PET image reconstruction is challenging due to the ill-poseness of the inverse problem and limited number of detected photons. Recently deep neural networks have been widely and successfully used in computer vision tasks and attracted growing interests in medical imaging. In this work, we trained a deep residual convolutional neural network to improve PET image quality by using the existing inter-patient information. An innovative feature of the proposed method is that we embed the neural network in the iterative reconstruction framework for image representation, rather than using it as a post-processing tool. We formulate the objective function as a constraint optimization problem and solve it using the alternating direction method of multipliers (ADMM) algorithm. Both simulation data and hybrid real data are used to evaluate the proposed method. Quantification results show that our proposed iterative neural network method can outperform the neural network denoising and conventional penalized maximum likelihood methods.