This is perhaps a very niche request, but if anyone has any similar experience I’d value any advice!
I am currently setting up my postprocessing pipeline in XCP-D. I’m working with nifti files, using the nichart Mode. I have specific timepoints in my data that (for various reasons) I want interpolated, but which don’t necessary have FD values above my FD threshold. I am wondering if there is a way I can create a custom column in my confounds file that includes all my FD values and artificially inflated FD values for the specific timepoints I want interpolated. Would it be possible to point XCP-D to this custom column, rather than the FD column, for identifying high motion timepoints? Specifically I’m interested in pointing XCP-D to custom columns to interpolate/censor, NOT as custom confounds for regression. Is this possible to do?
If it’s helpful to know, to retain the the temporal structure of my data, I replace timpoints over my FD thresold with NaNs, rather than censoring in XCP-D (which removes the timepoints entirely). So these interpolated timepoints will eventually be removed from the data, with the timing structure retained.