Nipype FramewiseDisplacement and tSNR

1

Dear all,

1.) tSNR: How do I generate a mean tSNR value? ( I get a mean.nii.gz file)

2.) FD:
I want to calculate my FD and use the following commands:

fd = FramewiseDisplacement()

fd.inputs.in_file = str(subject_path)+’/derivatives/participant/fmriprep/sub-’+str(subject_label)+’/ses-’+str(k)+’/func/sub-’+str(subject_label)+’_ses-’+str(k)+’_task-rest_run-3_space-MNI152NLin2009cAsym_desc-preproc_bold.nii.gz’

fd.inputs.parameter_source = ‘AFNI’

fd_result = fd.run()

which results in the following error:
UnicodeDecodeError: ‘utf-8’ codec can’t decode byte 0xf8 in position 48: invalid start byte

3.) I also want to calculate FD and tSNR after I regressed out ICA-Aroma noise components and GS, csf, wm and did filtering and smoothing etc. But I do not get any values which make sense.

What do I need to change?

2

It looks like you ran fMRIPrep. FD is one of the columns in the confounds.tsv file. It is a property of the run itself, so it is the same whether you use it in subject space, MNI9 (fmriprep default), or ICA aroma (MNI6) space.

I don’t have any guidance about tSNR.

3

I thought I could estimate the improvement of the FD after running fMRIPrep and further preprocessing steps? So I wanted to estimate the FD of the final file after running fMRIPrep, nuisance regression etc.

4

The FD is a property of how much the subject moved while being scanned. fMRIPrep will align all images to the same space, so there won’t be any “movement” in preprocessed images.