Moner: Motion Correction in Undersampled Radial MRI with Unsupervised Neural Representation

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Wu, Qing, Du, Chenhe, Tian, Xuanyu, Yu, Jingyi, Zhang, Yuyao, Wei, Hongjiang
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866912483696967680
author Wu, Qing
Du, Chenhe
Tian, Xuanyu
Yu, Jingyi
Zhang, Yuyao
Wei, Hongjiang
author_facet Wu, Qing
Du, Chenhe
Tian, Xuanyu
Yu, Jingyi
Zhang, Yuyao
Wei, Hongjiang
contents Motion correction (MoCo) in radial MRI is a particularly challenging problem due to the unpredictability of subject movement. Current state-of-the-art (SOTA) MoCo algorithms often rely on extensive high-quality MR images to pre-train neural networks, which constrains the solution space and leads to outstanding image reconstruction results. However, the need for large-scale datasets significantly increases costs and limits model generalization. In this work, we propose Moner, an unsupervised MoCo method that jointly reconstructs artifact-free MR images and estimates accurate motion from undersampled, rigid motion-corrupted k-space data, without requiring any training data. Our core idea is to leverage the continuous prior of implicit neural representation (INR) to constrain this ill-posed inverse problem, facilitating optimal solutions. Specifically, we integrate a quasi-static motion model into the INR, granting its ability to correct subject's motion. To stabilize model optimization, we reformulate radial MRI reconstruction as a back-projection problem using the Fourier-slice theorem. Additionally, we propose a novel coarse-to-fine hash encoding strategy, significantly enhancing MoCo accuracy. Experiments on multiple MRI datasets show our Moner achieves performance comparable to SOTA MoCo techniques on in-domain data, while demonstrating significant improvements on out-of-domain data. The code is available at: https://github.com/iwuqing/Moner
format Preprint
id arxiv_https___arxiv_org_abs_2409_16921
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Moner: Motion Correction in Undersampled Radial MRI with Unsupervised Neural Representation
Wu, Qing
Du, Chenhe
Tian, Xuanyu
Yu, Jingyi
Zhang, Yuyao
Wei, Hongjiang
Image and Video Processing
Computer Vision and Pattern Recognition
Motion correction (MoCo) in radial MRI is a particularly challenging problem due to the unpredictability of subject movement. Current state-of-the-art (SOTA) MoCo algorithms often rely on extensive high-quality MR images to pre-train neural networks, which constrains the solution space and leads to outstanding image reconstruction results. However, the need for large-scale datasets significantly increases costs and limits model generalization. In this work, we propose Moner, an unsupervised MoCo method that jointly reconstructs artifact-free MR images and estimates accurate motion from undersampled, rigid motion-corrupted k-space data, without requiring any training data. Our core idea is to leverage the continuous prior of implicit neural representation (INR) to constrain this ill-posed inverse problem, facilitating optimal solutions. Specifically, we integrate a quasi-static motion model into the INR, granting its ability to correct subject's motion. To stabilize model optimization, we reformulate radial MRI reconstruction as a back-projection problem using the Fourier-slice theorem. Additionally, we propose a novel coarse-to-fine hash encoding strategy, significantly enhancing MoCo accuracy. Experiments on multiple MRI datasets show our Moner achieves performance comparable to SOTA MoCo techniques on in-domain data, while demonstrating significant improvements on out-of-domain data. The code is available at: https://github.com/iwuqing/Moner
title Moner: Motion Correction in Undersampled Radial MRI with Unsupervised Neural Representation
topic Image and Video Processing
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2409.16921