Preprocessed BOLD fMRI time series in same space as full Freesurfer segmentation (aseg.mgz)

Software Support
1

Hi,

I am new to fMRIPrep and am having some trouble navigating the output files. For my project, I want to interpolate a BOLD signal onto a brain mesh generated from a T1 image with the end goal of using it as a source term in a FEM-based pulsatility simulation.

To create the mesh, I need a segmentation including ventricles and CSF space (freesurfer labels 4, 14, 15, 24, 43) . The file fmriprep/sourcedata/freesurfer/sub-01/mri/aseg.mgz has what I need. I cannot find any files under fmriprep/sub-01/anat that has all necessary labels.

Next, I need the full preprocessed BOLD time series in the same physical space as the segmentation I use for meshing. Looking under fmriprep/sub-01/func/ I find

sub-01_task-checker_run-01_desc-confounds_timeseries.json
sub-01_task-checker_run-01_desc-confounds_timeseries.tsv
sub-01_task-checker_run-01_desc-coreg_boldref.json
sub-01_task-checker_run-01_desc-coreg_boldref.nii.gz
sub-01_task-checker_run-01_desc-hmc_boldref.json
sub-01_task-checker_run-01_desc-hmc_boldref.nii.gz
sub-01_task-checker_run-01_from-boldref_to-T1w_mode-image_desc-coreg_xfm.json
sub-01_task-checker_run-01_from-boldref_to-T1w_mode-image_desc-coreg_xfm.txt
sub-01_task-checker_run-01_from-orig_to-boldref_mode-image_desc-hmc_xfm.json
sub-01_task-checker_run-01_from-orig_to-boldref_mode-image_desc-hmc_xfm.txt
sub-01_task-checker_run-01_space-MNI152NLin2009cAsym_boldref.nii.gz
sub-01_task-checker_run-01_space-MNI152NLin2009cAsym_desc-brain_mask.nii.gz
sub-01_task-checker_run-01_space-MNI152NLin2009cAsym_desc-preproc_bold.json
sub-01_task-checker_run-01_space-MNI152NLin2009cAsym_desc-preproc_bold.nii.gz

Among these, at least sub-01_task-checker_run-01_space-MNI152NLin2009cAsym_desc-preproc_bold.nii.gz has the full time series, so I may be able to use that if I get the segmentation to the same space.

I think I need to either:
a) transform the aseg.mgz into MNI space and use the <...>space-MNI152NLin2009cAsym_desc-preproc_bold.nii.gz file. Under fmriprep/sub-01/anat I find what I assume are transforms,

sub-01_from-MNI152NLin2009cAsym_to-T1w_mode-image_xfm.h5	
sub-01_from-T1w_to-MNI152NLin2009cAsym_mode-image_xfm.h5

but cannot figure out how to apply them.
b) re-run fmriprep with some additional options to generate the outputs I need automatically.

Appreciate any advice!

Thanks,
Ingvild

2

Hi @ingvildsd, and welcome to neurostars!

Perhaps the --output-space fsnative would be useful for you?

From Defining standard and nonstandard spaces where data will be resampled — fmriprep version documentation :

  • fsnative: similarly to the anat space for volumetric references, including the fsnative space will instruct fMRIPrep to sample the original BOLD data onto FreeSurfer’s reconstructed surfaces for this individual.

Best,
Steven

3

Thanks, @Steven! I’ll have a look at that.