Fmriprep: pick middle of ribbon when doing EPI-to-surface projection

Dear fmriprep experts,

I found that when fmriprep does projection from EPI to fsaverage space, it equally samples 6 points between pial surface and white-matter surface and averages them. But instead, if I want to only pick the middle point of the ribbon, is there a way to modify fmriprep code to achieve this?

This line of code (https://github.com/poldracklab/fmriprep/blob/bbbb2a81734c4eb16331bbb08cb5f3b012e7a358/fmriprep/workflows/bold/resampling.py#L124) seems doing it. Can I achieve my goal by simply changing this line to be:

fs.SampleToSurface(sampling_method=‘average’, sampling_range=(0.5),
sampling_units=‘frac’, interp_method=‘trilinear’, cortex_mask=True,
override_reg_subj=True, out_type=‘gii’),

Thank you!

Regards,
Jingwei

You’ll also need to change sampling_method to 'point', but yes that should work.

Thanks, Chris. I will try and see how it works. :slightly_smiling_face:

Best,
Jingwei

@Jingwei_Li hat’s the justification for choosing one projection strategy over the other?

We are going to explore that. The reason why we consider that is by using our own preprocessing pipeline and perform Yeo2011 parcellation algorithm, we got auditory/motor parcels split, but we didn’t replicate this auditory-motor split when using fmriprep on the same set of subjects. We are guessing there might be some information leakage through cortical ribbon.

Interesting - please share what you will find!