We are conducting a 2D CSI MRS study on acutely ill and recovered anorexia nervosa patients as well as healthy control participants (scanner SIEMENS PRISMA 3T). We positioned our VOI (8x8 voxels) directly above the corpus callosum and would like to do group comparisons cross-sectionally across our three groups and longitudinally in the acute anorexia nervosa group. We are aiming to do a group comparison within grey matter and white matter (entire VOI) as well as in the ROI ACC.

We would like to design a workflow for a quality check of our MRS data and we have two unresolved questions:

Would you recommend a quality check on voxel level or rather for the entire VOI?

Cramér-Rao lower bounds (CRLB) and FWHM are often used to assess spectra quality. If the CRLB are too high, would you recommend to exclude the entire voxel or just the respective metabolite?

All input is very much appreciated!