Comparison of linear combination modeling strategies for edited magnetic resonance spectroscopy at 3 T

Abstract

J-difference-edited spectroscopy is a valuable approach for the in vivo detection of γ-aminobutyric-acid (GABA) with magnetic resonance spectroscopy (MRS). A recent expert consensus article recommends linear combination modeling (LCM) of edited MRS but does not give specific details regarding implementation. This study explores different modeling strategies to adapt LCM for GABA-edited MRS. Sixty-one medial parietal lobe GABA-edited MEGA-PRESS spectra from a recent 3-T multisite study were modeled using 102 different strategies combining six different approaches to account for co-edited macromolecules (MMs), three modeling ranges, three baseline knot spacings, and the use of basis sets with or without homocarnosine. The resulting GABA and GABA+ estimates (quantified relative to total creatine), the residuals at different ranges, standard deviations and coefficients of variation (CVs), and Akaike information criteria, were used to evaluate the models' performance. Significantly different GABA+ and GABA estimates were found when a well-parameterized MM 3co basis function was included in the model. The mean GABA estimates were significantly lower when modeling MM 3co , while the CVs were similar. A sparser spline knot spacing led to lower variation in the GABA and GABA+ estimates, and a narrower modeling range-only including the signals of interest-did not substantially improve or degrade modeling performance. Additionally, the results suggest that LCM can separate GABA and the underlying co-edited MM 3co . Incorporating homocarnosine into the modeling did not significantly improve variance in GABA+ estimates. In conclusion, GABA-edited MRS is most appropriately quantified by LCM with a wellparameterized co-edited MM 3co basis function with a constraint to the nonoverlapped MM 0.93 , in combination with a sparse spline knot spacing (0.55 ppm) and a modeling range of 0.5-4 ppm.

Comparison of seven modelling algorithms for γ‐aminobutyric acid–edited proton magnetic resonance spectroscopy

“…A variation of this basis set was created, incorporating an additional Gaussian component at 3.0 ppm (simulated with FWHM = 14 Hz and scaled intensity equivalent to two protons) to represent coedited macromolecule signal underlying the GABA peak around 3.0 ppm. The amplitude scaling is consistent with the assumptions of a pseudo-doublet GABA signal at 3 ppm, and around 50% macromolecule contribution [38][39][40][41][42] to the observed signal in that area; the FWHM parameter had been optimised previously 43 on an aggregate subset of the data, over the 1-20 Hz range. This component, denoted MM3co, allowed the influence of macromolecule modelling on the various algorithms to be examined; subsequent use of this basis set is annotated with +MM3.…”

“…This component, denoted MM3co, allowed the influence of macromolecule modelling on the various algorithms to be examined; subsequent use of this basis set is annotated with +MM3. The interaction of this component with baseline stiffness and soft constraint models similar to those of [43][44][45] is explored for Osprey and LCModel in the supporting information, section B. All basis set algorithms were run both with and without the MM3 component; in all cases, the reported GABA+ values include contributions from the underlying macromolecule signal, either explicitly in cases where the MM3 component was modelled (i.e., GABA + MM3co), or implicitly in cases where it was not.…”

“…This may give rise to large residuals or biased baseline estimates if not considered in the model. Although their impact has been studied by some, 39,41,[43][44][45]73 there is currently no consensus on how these should be handled.…”

Importance of Linear Combination Modeling for Quantification of Glutathione and γ-Aminobutyric Acid Levels Using Hadamard-Edited Magnetic Resonance Spectroscopy

“…However, those comparisons would be overwhelming and beyond the scope of a single study. Thirdly, it is notable that the within-subject CV of GABA was higher than seen on average for between-subject CVs in a recent multi-site GABA+ study ( 47 ). This may reflect the limited estimation of variance in a single study with a low sample size (and these values are within the range of prior work), but also the different acquisition parameters.…”

Importance of linear combination modeling for quantification of Glutathione and GABA levels using Hadamard-edited MRS

“…However, those comparisons would be overwhelming and beyond the scope of a single study. Thirdly, it is notable that the within-subject CV of GABA was higher than seen on average for between-subject CVs in a recent multi-site GABA+ study 39 . This may reflect the limited estimation of variance in a single study with a low sample size (and these values are within the range of prior work), but also the different acquisition parameters.…”