Logo image
Back
Constrained Magnetic Resonance Spectroscopic Imaging by Learning Nonlinear Low-Dimensional Models
Journal article   Open access   Peer reviewed

Constrained Magnetic Resonance Spectroscopic Imaging by Learning Nonlinear Low-Dimensional Models

Fan Lam, Yahang Li and Xi Peng
IEEE transactions on medical imaging, Vol.39(3), pp.545-555
03/2020
DOI: 10.1109/TMI.2019.2930586
PMID: 31352337
url
https://doi.org/10.1109/TMI.2019.2930586View
Published (Version of record) Open Access

Abstract

Magnetic resonance spectroscopic imaging (MRSI) is a powerful molecular imaging modality but has very limited speed, resolution, and SNR tradeoffs. Construction of a low-dimensional model to effectively reduce the dimensionality of the imaging problem has recently shown great promise in improving these tradeoffs. This paper presents a new approach to model and reconstruct the spectroscopic signals by learning a nonlinear low-dimensional representation of the general MR spectra. Specifically, we trained a deep neural network to capture the low-dimensional manifold, where the high-dimensional spectroscopic signals reside. A regularization formulation is proposed to effectively integrate the learned model and physics-based data acquisition model for MRSI reconstruction with the capability to incorporate additional spatiospectral constraints. An efficient numerical algorithm was developed to solve the associated optimization problem involving back-propagating the trained network. Simulation and experimental results were obtained to demonstrate the representation power of the learned model and the ability of the proposed formulation in producing SNR-enhancing reconstruction from the practical MRSI data.
 Spectroscopy Data models Feature extraction Image reconstruction Imaging low-dimensionalmodels manifold learning Manifolds MR spectroscopic imaging neural network Neural networks Signal to noise ratio spatiospectral constraint

Details

Metrics

15 Record Views
41 Times Cited - Web of Science
Logo image