Learning Bloch Simulations for MR Fingerprinting by Invertible Neural Networks

Magnetic resonance fingerprinting (MRF) enables fast and multiparametric MR imaging. Despite fast acquisition, the state-of-the-art reconstruction of MRF based on dictionary matching is slow and lacks scalability. To overcome these limitations, neural network (NN) approaches estimating MR parameters...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2021-03
Main Authors: Balsiger, Fabian, Jungo, Alain, Scheidegger, Olivier, Marty, Benjamin, Reyes, Mauricio
Format: Article
Language:English
Subjects:
Computer simulation
Fingerprinting
Fingerprints
Learning
Magnetic resonance imaging
Neural networks
Parameter estimation
Process parameters
Reconstruction
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Magnetic resonance fingerprinting (MRF) enables fast and multiparametric MR imaging. Despite fast acquisition, the state-of-the-art reconstruction of MRF based on dictionary matching is slow and lacks scalability. To overcome these limitations, neural network (NN) approaches estimating MR parameters from fingerprints have been proposed recently. Here, we revisit NN-based MRF reconstruction to jointly learn the forward process from MR parameters to fingerprints and the backward process from fingerprints to MR parameters by leveraging invertible neural networks (INNs). As a proof-of-concept, we perform various experiments showing the benefit of learning the forward process, i.e., the Bloch simulations, for improved MR parameter estimation. The benefit especially accentuates when MR parameter estimation is difficult due to MR physical restrictions. Therefore, INNs might be a feasible alternative to the current solely backward-based NNs for MRF reconstruction.
ISSN:2331-8422
DOI:10.48550/arxiv.2008.04139