fMRIVolume pipeline: steps after minimal preprocessing
Matthieu Vanhoutte
- hp2000: high-pass filtering
- clean: sICA+FIX cleaning
- rclean: Is this motion-regression? If this is the case, wasn't it prohibited before? Where are regressors data?
- tclean:Is this tICA cleaning?
Glasser, Matthew
The recommended file to use is ${StudyFolder}/${Subject}/MNINonLinear/Results/rfMRI_REST_MSMAll_hp2000_clean_rclean_tclean.dtseries.nii.
This CIFTI grayodinates file has:
- Been aligned across individuals on the cortical surface using multi-modal information including functional information so that cortical areas and functional networks are properly aligned. This is discussed in Glasser et al., 2016 Nature (the particular version used in HCP-YA).
- Cleaned with spatial ICA denoising to remove spatially specific artifacts (e.g., from head motion). _clean_rclean means that an improved classification was done of the spatial ICA components. New from before, these data do not have movement regressors removed, because that was found to be deleterious (Glasser et al., 2019 Neuroimage).
- Clean with temporal ICA denoising to remove global artifacts (e.g., from respiration variation). This is discussed in Glasser et al., 2018 Neuroimage.
NIFTI volume data are not recommended for primary analyses. There are cases in which it is helpful to have both CIFTI and volume outputs, but those are rarer. The reasons for this are discussed in Coalson et al., 2018 PNAS and Glasser et al., 2016 Nature Neuroscience.
Matt.
--
You received this message because you are subscribed to the Google Groups "HCP-Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to
hcp-users+...@humanconnectome.org.
To view this discussion visit
https://groups.google.com/a/humanconnectome.org/d/msgid/hcp-users/CAF6uKJQCYPRyTzxV2eupyoQviiFzJWoyL7kKkZVYsX3y4f4DXg%40mail.gmail.com.
The materials in this message are private and may contain Protected Healthcare Information or other information of a sensitive nature. If you are not the intended recipient, be advised that any unauthorized use, disclosure, copying or the taking of any action in reliance on the contents of this information is strictly prohibited. If you have received this email in error, please immediately notify the sender via telephone or return mail.
Matthieu Vanhoutte
Glasser, Matthew
There isn’t a good way to do that for the same reason that we don’t recommend volume-based analysis of the cerebral cortex in humans. Please see Coalson et al., 2018 PNAS for why.
Matthieu Vanhoutte
Glasser, Matthew
There is no way to project a group volume atlas to the surface that is as accurate as if your group atlas were properly made on the surface. The problem is with the group volume atlas, not with the grayordinates or surface files.