WKGM: Weight-K-space Generative Model for Parallel Imaging Reconstruction

05/08/2022
by   Zongjiang Tu, et al.
5

Parallel Imaging (PI) is one of the most im-portant and successful developments in accelerating magnetic resonance imaging (MRI). Recently deep learning PI has emerged as an effective technique to accelerate MRI. Nevertheless, most approaches have so far been based image domain. In this work, we propose to explore the k-space domain via robust generative modeling for flexible PI reconstruction, coined weight-k-space generative model (WKGM). Specifically, WKGM is a generalized k-space domain model, where the k-space weighting technology and high-dimensional space strategy are efficiently incorporated for score-based generative model training, resulting in good and robust reconstruction. In addition, WKGM is flexible and thus can synergistically combine various traditional k-space PI models, generating learning-based priors to produce high-fidelity reconstructions. Experimental results on datasets with varying sampling patterns and acceleration factors demonstrate that WKGM can attain state-of-the-art reconstruction results under the well-learned k-space generative prior.

READ FULL TEXT

page 1

page 3

page 4

page 5

page 7

page 8

page 9

page 10

research
08/15/2022

One-shot Generative Prior Learned from Hankel-k-space for Parallel Imaging Reconstruction

Magnetic resonance imaging serves as an essential tool for clinical diag...
research
08/14/2020

Homotopic Gradients of Generative Density Priors for MR Image Reconstruction

Deep learning, particularly the generative model, has demonstrated treme...
research
03/21/2022

K-space and Image Domain Collaborative Energy based Model for Parallel MRI Reconstruction

Decreasing magnetic resonance (MR) image acquisition times can potential...
research
04/05/2022

Multi-Weight Respecification of Scan-specific Learning for Parallel Imaging

Parallel imaging is widely used in magnetic resonance imaging as an acce...
research
01/19/2022

Virtual Coil Augmentation Technology for MRI via Deep Learning

Magnetic resonance imaging (MRI) is a widely used medical imaging modali...
research
09/07/2021

MRI Reconstruction Using Deep Energy-Based Model

Purpose: Although recent deep energy-based generative models (EBMs) have...
research
12/15/2022

Universal Generative Modeling in Dual-domain for Dynamic MR Imaging

Dynamic magnetic resonance image reconstruction from incomplete k-space ...

Please sign up or login with your details

Forgot password? Click here to reset