fMRI analysis

[Mudathir Bakhit]

Good evening everyone,

I hope all of you are safe and doing well on these difficult days.

If you allow me, I have few questions regarding the fMRI analysis of the HCP data.

I read few threads in this forum and understood (please correct me if I am wrong) that regarding rfMRI seed-based analysis, at the individual level, it is recommended to use the Atlas_MSMAll_hp2000_clean.dtseries data, and at the average level is the groupPCA_d4500_Eigenmaps.dtseries data by applying the wb_command -cifti-average-roi-correlation command on both.

My question is how to import a binary ROI drawn in fsleyes to wb_command for seed-based analysis for rfMRI or ROI analysis in tfMRI?

Next question, if succeeded and imported the ROI and did the seed-based analysis, how to create a temporal concatenation of these files, so each file looks like a separate timepoint in a 4d file? I have ROI1, ROI2, …ROI10, and each used to produce a dscalar file. I want to do a temporal concatenation of these files. Then, how to cross-correlate two time-series in wb_command? In fsl, this can be done by fslcc, is there a similar command in wb_commands? I mean a correlation of the temporal concatenation file vs. HCP ICA parcellations (e.g., 3T_HCP1200_MSMAll_d15_ts2_Z/ 105115.dtseries.nii)

Last question, regarding the tfMRI, what is the recommended average template to use in ROI analysis?

Thanks

Mudathir

[Glasser, Matthew]

1. Is this a surface or volume ROI?  Group or individual?
2. You can concatenate files with wb_command -cifti-merge
3. What kind of analysis are you wanting to do with tfMRI?

 

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 on the web visit https://groups.google.com/a/humanconnectome.org/d/msgid/hcp-users/55007f6b-e36f-4af9-ba8d-f84898926819n%40humanconnectome.org.

 

---

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.

[Mudathir Bakhit]

Thank you Matt for the quick replay. I appreciate it.

1. It is a binary volume in MNI space. I have both, for each individual ROI and as well a group ROI (average). because we plan to do both individual and group analysis.

3. with tfMRI I want to do an ROI analysis to measure the average statistical value (e.g., z score) in subjects or group. In the subject I already thought about using featquery of the volume cope.feat data.

Mudathir

[Glasser, Matthew]

It would be challenging to do an accurate analysis with group volume ROIs in the cerebral cortex.   It is workable subcortically though and in individual subjects (assuming the ROIs are limited to the cortical ribbon).

[Mudathir Bakhit]

Yes, after resampling we confirmed the ROIs are satisfactorily within the cortical ribbon in each subject we studied. So I guess it can work with individual analysis.

Mudathir

[Coalson, Timothy Scott (S&T-Student)]

If you still want to additionally do a group-based method, you could draw closed borders on the surface in wb_view (typically on an inflated surface, using myelin and/or some fMRI results as a guide), and convert those to surface rois with wb_command -border-to-rois.  Surface-based processing and MSMAll registration gets around the major issues with volume-based group average cortex.

Tim

---

From: Mudathir Bakhit <mudat...@gmail.com>
Sent: Sunday, January 24, 2021 7:18 AM
To: HCP-Users <hcp-...@humanconnectome.org>
Cc: glas...@wustl.edu <glas...@wustl.edu>

To view this discussion on the web visit https://groups.google.com/a/humanconnectome.org/d/msgid/hcp-users/5e634c38-b706-424c-a68e-612a67527a1en%40humanconnectome.org.