Model-Informed Unsupervised Deep Learning Approaches to Frequency and Phase Correction of MRS Signals
Preprint
Abstract
Purpose A supervised deep learning (DL) approach for frequency-and-phase Correction (FPC) of MR spectroscopy (MRS) data recently showed encouraging results, but obtaining transients with labels for supervised learning is challenging. This work investigates the feasibility and efficiency of unsupervised DL-based FPC. Method Two novel DL-based FPC methods (deep learning-based Cr referencing [dCrR] and deep learning-based spectral registration [dSR]) which use a priori physics domain knowledge are presented. The proposed networks were trained, validated, and evaluated using simulated, phantom, and publicly accessible in-vivo MEGA-edited MRS data. The performance of our proposed FPC methods was compared to other generally used FPC methods, in terms of precision and time efficiency. A new measure was proposed in this study to evaluate the FPC method performance. The ability of each of our methods to carry out FPC at varying SNR levels was evaluated. A Monte Carlo (MC) study was carried out to investigate the performance of our proposed methods. Result The validation using low-SNR manipulated simulated data demonstrated that the proposed methods could perform FPC comparably to other methods. The evaluation showed that the dCrR method achieved the highest performance in phantom data. The applicability of the proposed method for FPC of GABA-edited in-vivo MRS data was demonstrated. Our proposed networks have the potential to reduce computation time significantly. Conclusion The proposed physics-informed deep neural networks trained in an unsupervised manner with complex data can offer efficient FPC of MRS data in a shorter time.Keywords
All Related Versions
- Published version: Magnetic Resonance in Medicine, 89 (3), 1221.
This publication has 32 references indexed in Scilit:
- Advanced processing and simulation of MRS data using the FID appliance (FID‐A)—An open source, MATLAB‐based toolkitMagnetic Resonance in Medicine, 2015
- Frequency and phase drift correction of magnetic resonance spectroscopy data by spectral registration in the time domainMagnetic Resonance in Medicine, 2014
- Gannet: A batch‐processing tool for the quantitative analysis of gamma‐aminobutyric acid–edited MR spectroscopy spectraJournal of Magnetic Resonance Imaging, 2013
- A novel phase and frequency navigator for proton magnetic resonance spectroscopy using water‐suppression cyclingMagnetic Resonance in Medicine, 2010
- Prospective motion correction for single‐voxel 1H MR spectroscopyMagnetic Resonance in Medicine, 2010
- A practical guide to robust detection of GABA in human brain by J-difference spectroscopy at 3 T using a standard volume coilMagnetic Resonance Imaging, 2007
- Reducing the Dimensionality of Data with Neural NetworksScience, 2006
- Linear low-rank approximation and nonlinear dimensionality reductionScience China Mathematics, 2004
- Restoration of motion‐related signal loss and line‐shape deterioration of proton MR spectra using the residual water as intrinsic referenceMagnetic Resonance in Medicine, 2001
- Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in positionBiological Cybernetics, 1980