rs fMRI nifti volumetric files usage (HCP-Ageing V1)

[Mathis Macé]

Hi, 

I recently downloaded HCP-A V1 recommended preprocessed rs-fMRI data (a day before the release of V2!) and started using the recommended dtseries.nii files. But their CIFTI format and associated dimensions are new to me and quite different compared to the other dataset's formats I'm currently analyzing. 

There is also the file : rfMRI_REST_hp0_clean_rclean_tclean.nii.gz, which format I'm more familiar with, on which I think I could use the same exact analysis pipelines as in other datasets.

But, in the AABC_Release1_Imaging_Data_Packages.pdf doc, it's said, "This package contains recommended rfMRI files (for all resting-state runs) including spatial and temporal ICA-cleaned CIFTI files precisely aligned across subjects using the MSMAll multi-modal surface registration. It also contains NIFTI volumetric files that in some cases are appropriate to analyze together with the CIFTI files (e.g., to compare the two). Analyzing NIFTI volumetric files alone is NOT RECOMMENDED" so I was wondering why it is not recommended ? What are the specific reasons here? Because as it is, it appears to me that this file could be more appropriate in terms of analysis comparability and continuity for my research project. 

Thanks in advance for your help.

Best regards,

Mathis

[Tim Coalson]

Group analysis of human cortex is made difficult by the differences in folding geometry between subjects.  Surface-based analysis enables precise group analysis of this data by extracting each subjects' cortical data using their own segmentations before any group averaging is done (which also enables finding a more precise correspondence between cortical features across subjects), while volume-based group registration and analysis only gives fuzzy answers that, among other things, often conflate opposing sulcal banks due to their proximity.  See our paper on this topic:

https://www.pnas.org/doi/10.1073/pnas.1801582115

The cifti-format files represent the (cerebral) cortical data using surface vertices, and subcortical data using voxels, so all of the gray matter can be analyzed together using the appropriate representations.

Tim

--
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/383365f3-a515-4d58-b8e7-0c616471b872n%40humanconnectome.org.

[Glasser, Matthew]

I would add that doing high quality science sometimes requires going outside of your comfort zone.  Scientific inertia is one of the biggest barriers to scientific progress.

Matt.

To view this discussion visit https://groups.google.com/a/humanconnectome.org/d/msgid/hcp-users/CANm3QU6wUL6%2B0YMZVjiCBgz5V6VPtnq6CXGcSALqdLh3FkeU6w%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.