Recent Advances in in-vivo Spectroscopy at CNI

Recent Advances in in-vivo Spectroscopy Methods

Background

CNI continues to support, via the CNI spectroscopy special interest group, the research of its user community.  This support includes developing and incorporating for general use new data acquisition and data analysis capabilities as recommended by the at-large MRS community. The primary focus is on measuring metabolic changes via MRS techniques and combining that information with functional MRI measurements in neuroscience studies. If you are interested in learning more about these methods, or joining the special interest group, please consult with Laima Baltusis.

The most recent spectroscopy methods developments and results were shared in a poster presented at the 66^th joint ENC-ISMAR Conference, the premier conference for nuclear magnetic research (April 6 – 10, 2025). The key technical points in the poster are summarized below. Details and supporting figures are in the poster attached below.

Technical Notes

For all spectroscopy research projects CNI actively follows the recommended best practices from the ISMRM (International Society for Magnetic Resonance in Medicine) Spectroscopy Study Group and recently published experts’ consensus recommendations. Currently the semi-LASER (sLASER) sequence is the ISMRM consensus method, particularly for multi-site 3T MRS studies. We have, therefore, transitioned studies at CNI from previously recommended sequences (PRESS, as example) to the sLASER sequence (a GE product sequence) for both short TE single voxel and new focal 2D MRSI studies.

To measure and improve data quality of small metabolite differences in brain regions and subregions of interest, we have been optimizing and evaluating focal 2D MRSI data acquisition in both cortical and subcortical regions of the brain as an alternative to single voxel MRS data acquisition with the sLASER sequence. Typically, single voxel data particularly in the subcortical regions has been seen to be of lower quality compared to other brain regions. The subcortical brain regions are important and have been often studied by investigators trying to understand movement, reward, cognition, and emotion.

Data quality of the focal 2D MRSI in the subcortical region of the brain can be seen to be superior compared to single voxel MRS data acquisition. Our results indicate fewer artifacts, fewer baseline issues, and better signal-to-noise in focal 2D MRSI data relative to single voxel MRS data as input for data analysis. Additionally, MRSI data from the subcortical Basal Ganglia regions also demonstrate the flexibility to separate different Basal Ganglia (BG) regions of interest such as Caudate, Putamen, and Globus Pallidus as well as other regions of interest including Insula and Thalamus.

Initial basic data reconstruction and visualization employed SAGE, a GE proprietary analysis and visualization tool with LCModel used as a separate module for data fitting. We have now moved our data reconstruction and visualization to open-source in-house MATLAB scripts. The key steps for data reconstruction, visualization, and fitting are outlined below.

Reconstruction:

• Suppression of leakage of the artifact signals from voxels outside of the PRESS box, with a 2D version of the spatial Fermi filter is applied to the k-space data before Fourier transform.

• Coil combination performed voxel by voxel using summed square of the magnitude of the first data point in the FID.

• The reconstructed spectral grid is extrapolated to a grid of 32×32 before LCModel quantification.

Visualization:

• A regional MRSI mask created from the excited volume.

• The MRSI mask co-registered to a FreeSurfer parcellated brain regions map using a T1-weighted MRI series, and the white and gray matter tissue contributions in every MRSI voxel calculated.

• The spectra from the voxels identified for a particular brain subregion by FreeSurfer parcellation (for example putamen) are averaged for metabolite quantification, and statistical analyses.

Data Fitting:

• LCModel is used as a separate module for data fitting.

• For optimal quantification of metabolites in both focal 2D MRSI and single voxel MRS data, the analysis methods include improvements in data analysis using LCModel by (1) mitigation of non-metabolite contributions for data analysis with LCModel and (2) improvement of the accuracy of the LCModel basis using a largely experimental 23 metabolite basis set.

• To correct for potential influences on neurometabolite estimates from different tissue amounts in each voxel, the estimates of percent gray matter in each voxel are used in analyses as covariates.

To develop data acquisition and data processing methodology for research studies, MRSI data were collected and analyzed pilot from N=7 healthy control participants (4 female, 3 male) each scanned twice.

Pilot data from these healthy control participants using test re-test methods demonstrated that measured neurometabolites to be highly reproducible within persons and between left and right subcortical Basal ganglia regions, thalamus, and insula.

The automated analysis methods developed with pilot control participants will now be applied to collected data of research participant populations to evaluate neurometabolite changes in subcortical regions and their subregions resulting from alcohol and substance abuse with metabolite profiles compared to healthy controls.

Future research directions will include correlating focal MRSI data in subcortical regions with methods such as QSM (iron susceptibility) and resting state fMRI (network connectivity) to determine if there is a relationship with metabolites on iron and/or brain networks.