Learning a preconditioner to accelerate compressed sensing reconstructions in MRI

Purpose To learn a preconditioner that accelerates parallel imaging (PI) and compressed sensing (CS) reconstructions. Methods A convolutional neural network (CNN) with residual connections was used to train a preconditioning operator. Training and validation data were simulated using 50% brain image...

Full description

Saved in:
Bibliographic Details
Published in:Magnetic resonance in medicine Vol. 87; no. 4; pp. 2063 - 2073
Main Authors: Koolstra, Kirsten, Remis, Rob
Format: Journal Article
Language:English
Published: United States Wiley Subscription Services, Inc 01.04.2022
John Wiley and Sons Inc
Subjects:
Acceleration
Algorithms
Artificial neural networks
Brain - diagnostic imaging
compressed sensing
Computer architecture
Conditioned stimulus
deep learning
Image Processing, Computer-Assisted - methods
Image reconstruction
Magnetic resonance imaging
Magnetic Resonance Imaging - methods
Medical imaging
MR image reconstruction
Neural networks
Neural Networks, Computer
Neuroimaging
parallel imaging
Parameter sensitivity
Preconditioning
Random noise
Regularization
Sampling
split Bregman
Technical Note
Technical Notes—Computer Processing and Modeling
Training
split Bregman
deep learning
parallel imaging
compressed sensing
MR image reconstruction
preconditioning
ISSN:0740-3194, 1522-2594, 1522-2594
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Purpose To learn a preconditioner that accelerates parallel imaging (PI) and compressed sensing (CS) reconstructions. Methods A convolutional neural network (CNN) with residual connections was used to train a preconditioning operator. Training and validation data were simulated using 50% brain images and 50% white Gaussian noise images. Each multichannel training example contains a simulated sampling mask, complex coil sensitivity maps, and two regularization parameter maps. The trained model was integrated in the preconditioned conjugate gradient (PCG) method as part of the split Bregman CS method. The acceleration performance was compared with that of a circulant PI‐CS preconditioner for varying undersampling factors, number of coil elements and anatomies. Results The learned preconditioner reduces the number of PCG iterations by a factor of 4, yielding a similar acceleration as an efficient circulant preconditioner. The method generalizes well to different sampling schemes, coil configurations and anatomies. Conclusion It is possible to learn adaptable preconditioners for PI and CS reconstructions that meet the performance of state‐of‐the‐art preconditioners. Further acceleration could be achieved by optimizing the network architecture and the training set. Such a preconditioner could also be integrated in fully learned reconstruction methods to accelerate the training process of unrolled networks.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:0740-3194
1522-2594
1522-2594
DOI:10.1002/mrm.29073