On learning adaptive acquisition policies for undersampled multi-coil MRI reconstruction

03/30/2022
by   Tim Bakker, et al.
0

Most current approaches to undersampled multi-coil MRI reconstruction focus on learning the reconstruction model for a fixed, equidistant acquisition trajectory. In this paper, we study the problem of joint learning of the reconstruction model together with acquisition policies. To this end, we extend the End-to-End Variational Network with learnable acquisition policies that can adapt to different data points. We validate our model on a coil-compressed version of the large scale undersampled multi-coil fastMRI dataset using two undersampling factors: 4× and 8×. Our experiments show on-par performance with the learnable non-adaptive and handcrafted equidistant strategies at 4×, and an observed improvement of more than 2% in SSIM at 8× acceleration, suggesting that potentially-adaptive k-space acquisition trajectories can improve reconstructed image quality for larger acceleration factors. However, and perhaps surprisingly, our best performing policies learn to be explicitly non-adaptive.

READ FULL TEXT

page 8

page 21

page 22

research
07/20/2020

Active MR k-space Sampling with Reinforcement Learning

Deep learning approaches have recently shown great promise in accelerati...
research
04/08/2022

A Learnable Variational Model for Joint Multimodal MRI Reconstruction and Synthesis

Generating multi-contrasts/modal MRI of the same anatomy enriches diagno...
research
10/27/2019

GrappaNet: Combining Parallel Imaging with Deep Learning for Multi-Coil MRI Reconstruction

Magnetic Resonance Image (MRI) acquisition is an inherently slow process...
research
08/10/2021

Optimal MRI Undersampling Patterns for Ultimate Benefit of Medical Vision Tasks

To accelerate MRI, the field of compressed sensing is traditionally conc...
research
02/09/2017

Incorporation of prior knowledge of the signal behavior into the reconstruction to accelerate the acquisition of MR diffusion data

Diffusion MRI measurements using hyperpolarized gases are generally acqu...
research
06/07/2020

A multi-channel framework for joint reconstruction of multi-contrast parallel MRI

Compressed sensing, multi-contrast and parallel imaging are all techniqu...
research
03/04/2021

Towards Ultrafast MRI via Extreme k-Space Undersampling and Superresolution

We went below the MRI acceleration factors (a.k.a., k-space undersamplin...

Please sign up or login with your details

Forgot password? Click here to reset