Dynamic PET Image Reconstruction via Non-negative INR Factorization

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Zhang, Chaozhi, Ding, Wenxiang, He, Roy Y., Zhang, Xiaoqun, Ding, Qiaoqiao
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909657864339456
author Zhang, Chaozhi
Ding, Wenxiang
He, Roy Y.
Zhang, Xiaoqun
Ding, Qiaoqiao
author_facet Zhang, Chaozhi
Ding, Wenxiang
He, Roy Y.
Zhang, Xiaoqun
Ding, Qiaoqiao
contents The reconstruction of dynamic positron emission tomography (PET) images from noisy projection data is a significant but challenging problem. In this paper, we introduce an unsupervised learning approach, Non-negative Implicit Neural Representation Factorization (\texttt{NINRF}), based on low rank matrix factorization of unknown images and employing neural networks to represent both coefficients and bases. Mathematically, we demonstrate that if a sequence of dynamic PET images satisfies a generalized non-negative low-rank property, it can be decomposed into a set of non-negative continuous functions varying in the temporal-spatial domain. This bridges the well-established non-negative matrix factorization (NMF) with continuous functions and we propose using implicit neural representations (INRs) to connect matrix with continuous functions. The neural network parameters are obtained by minimizing the KL divergence, with additional sparsity regularization on coefficients and bases. Extensive experiments on dynamic PET reconstruction with Poisson noise demonstrate the effectiveness of the proposed method compared to other methods, while giving continuous representations for object's detailed geometric features and regional concentration variation.
format Preprint
id arxiv_https___arxiv_org_abs_2503_08025
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Dynamic PET Image Reconstruction via Non-negative INR Factorization
Zhang, Chaozhi
Ding, Wenxiang
He, Roy Y.
Zhang, Xiaoqun
Ding, Qiaoqiao
Computer Vision and Pattern Recognition
The reconstruction of dynamic positron emission tomography (PET) images from noisy projection data is a significant but challenging problem. In this paper, we introduce an unsupervised learning approach, Non-negative Implicit Neural Representation Factorization (\texttt{NINRF}), based on low rank matrix factorization of unknown images and employing neural networks to represent both coefficients and bases. Mathematically, we demonstrate that if a sequence of dynamic PET images satisfies a generalized non-negative low-rank property, it can be decomposed into a set of non-negative continuous functions varying in the temporal-spatial domain. This bridges the well-established non-negative matrix factorization (NMF) with continuous functions and we propose using implicit neural representations (INRs) to connect matrix with continuous functions. The neural network parameters are obtained by minimizing the KL divergence, with additional sparsity regularization on coefficients and bases. Extensive experiments on dynamic PET reconstruction with Poisson noise demonstrate the effectiveness of the proposed method compared to other methods, while giving continuous representations for object's detailed geometric features and regional concentration variation.
title Dynamic PET Image Reconstruction via Non-negative INR Factorization
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2503.08025