Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Relaxometry Guided Quantitative Cardiac Magnetic Resonance Image Reconstruction

Yidong Zhao, Yi Zhang, Qian Tao

TL;DR

This work proposes a relaxometry-guided quantitative MRI reconstruction framework to learn the spatial prior from data and the relaxometry prior from MRI physics, and evaluated the performance of two popular reconstruction backbones, namely, recurrent variational networks (RVN) and variational networks with U- Net.

Abstract

Deep learning-based methods have achieved prestigious performance for magnetic resonance imaging (MRI) reconstruction, enabling fast imaging for many clinical applications. Previous methods employ convolutional networks to learn the image prior as the regularization term. In quantitative MRI, the physical model of nuclear magnetic resonance relaxometry is known, providing additional prior knowledge for image reconstruction. However, traditional reconstruction networks are limited to learning the spatial domain prior knowledge, ignoring the relaxometry prior. Therefore, we propose a relaxometry-guided quantitative MRI reconstruction framework to learn the spatial prior from data and the relaxometry prior from MRI physics. Additionally, we also evaluated the performance of two popular reconstruction backbones, namely, recurrent variational networks (RVN) and variational networks (VN) with U- Net. Experiments demonstrate that the proposed method achieves highly promising results in quantitative MRI reconstruction.

Relaxometry Guided Quantitative Cardiac Magnetic Resonance Image Reconstruction

TL;DR

This work proposes a relaxometry-guided quantitative MRI reconstruction framework to learn the spatial prior from data and the relaxometry prior from MRI physics, and evaluated the performance of two popular reconstruction backbones, namely, recurrent variational networks (RVN) and variational networks with U- Net.

Abstract

Deep learning-based methods have achieved prestigious performance for magnetic resonance imaging (MRI) reconstruction, enabling fast imaging for many clinical applications. Previous methods employ convolutional networks to learn the image prior as the regularization term. In quantitative MRI, the physical model of nuclear magnetic resonance relaxometry is known, providing additional prior knowledge for image reconstruction. However, traditional reconstruction networks are limited to learning the spatial domain prior knowledge, ignoring the relaxometry prior. Therefore, we propose a relaxometry-guided quantitative MRI reconstruction framework to learn the spatial prior from data and the relaxometry prior from MRI physics. Additionally, we also evaluated the performance of two popular reconstruction backbones, namely, recurrent variational networks (RVN) and variational networks (VN) with U- Net. Experiments demonstrate that the proposed method achieves highly promising results in quantitative MRI reconstruction.
Paper Structure (17 sections, 11 equations, 2 figures, 2 tables)

This paper contains 17 sections, 11 equations, 2 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Contributions
  3. Methods
  4. Parallel Imaging
  5. Accelerated MRI
  6. Sensitivity Estimation
  7. Relaxometry Guided Reconstruction
  8. Dual-domain Reconstruction Network
  9. Quantitative Mapping Network
  10. Joint Reconstruction and Quantitative Mapping
  11. Experiments
  12. Dataset
  13. Training Configuration
  14. Results
  15. Evaluation Results
  16. ...and 2 more sections

Figures (2)

  • Figure 1: The reconstruction backbone consists of unrolled gradient descent layers, and the image prior is learned during training by $\mathcal{G}_{\theta}$. A pre-trained mapping network $\mathcal{M}$ is introduced to predict the quantitative parameters $\bm{p}$ and guide the reconstruction with MR relaxometry.
  • Figure 2: Qualitative results for T1 and T2 mapping sequences. The baseline images with the shortest and longest inversion/echo times are shown. The proposed method can generate both images and quantitative maps simultaneously. Perceptually, the reconstructed images of all acceleration factors are of good quality.