N/A values in confounds tsv file

1

Hi all,

I am very new using fmriprep and conn. I ran fmriprep in a resting state dataset.
In the first line of the confounds.tsv file the framewise displacement confound has a N/A value and this scan is also detected as an outlier in column number_of_dummy_scans. When loading this confound into CONN I get an error. Can I just remove this first line or zero pad it? If so would it affect anything else? I used fMRIPrep 20.1.0rc3.

Any help is much appreciated.
Best,
Ana

2

Hi Ana,

For framewise displacement I believe replacing the first value (n/a) with the mean of the column is a valid solution based on another conversation.

So instead of zero padding, I would replace the n/a with the mean.

Best,
James

3

Thank you, James. That is very helpful!
Best,
Ana

1 Like