Regarding spatial smoothing of 7T movie-watching tfMRI data

[Bokyeong Jeong]

Hello,

I have a question regarding the volume-based time series data included in the HCP Young Adult dataset, specifically within the ConnectomeDB package “Movie Task fMRI FIX-Denoised (Extended)” (e.g., tfMRI_MOVIE1_7T_AP_hp2000_clean.nii). I am wondering whether this data has undergone spatial smoothing.

From my understanding based on the HCP_S1200_Reference_Manual(p.126-), the HCP minimal preprocessing pipeline for tfMRI data generally proceeds as follows: HCP Structural Pipeline (including PreFreeSurfer, FreeSurfer, and PostFreeSurfer steps) → HCP Functional Pipeline (including fMRIVolume and fMRISurface steps) → tfMRIDenoising (using ICA-FIX).

According to the manual, the fMRIVolume pipeline of the HCP Functional Pipeline includes “no overt volume smoothing done” (p.131), whereas the fMRISurface pipeline applies 2 mm FWHM spatial smoothing to the cortex and subcortical regions (p.132).

So my question is: for the volume-based time series data I downloaded, has it gone through both the fMRIVolume and fMRISurface pipelines(including 2mm minimal spatial smoothing included in this step) sequentially or has it been denoised using ICA-FIX directly after fMRIVolume (i.e., without the 2 mm spatial smoothing applied in the fMRISurface step)?

To investigate this, I tried using AFNI’s 3dFWHMx to estimate the spatial smoothness. However, I still couldn't determine whether the 2 mm minimal smoothing from the HCP preprocessing pipeline was applied. Below are the results I obtained:

----------
bokyeong@Spaceship:~$ 3dFWHMx \
 -input /mnt/d/Loneliness/131217/MNINonLinear/Results/tfMRI_MOVIE1_7T_AP/tfMRI_MOVIE1_7T_AP_hp2000_clean.nii \
 -mask  /mnt/d/Loneliness/131217/MNINonLinear/Results/tfMRI_MOVIE1_7T_AP/brainmask_fs.1.60.nii
++ 3dFWHMx: AFNI version=AFNI_25.2.09 (Aug 24 2025) [64-bit]
++ Authored by: The Bob
++ Number of voxels in mask = 441564
*+ WARNING: removed 144 voxels from mask because they are constant in time
++ start ACF calculations out to radius = 6.40 mm
 + ACF done (0.00 CPU s thus far)
 0  0  0    0
 0.956672  1.07155  8.30738    2.44532
++ ACF 1D file [radius ACF mixed_model gaussian_NEWmodel] written to 3dFWHMx.1D
++ 1dplot: AFNI version=AFNI_25.2.09 (Aug 24 2025) [64-bit]
++ Authored by: RWC et al.
pnmtopng: 40 colors found
 + and 1dplot-ed to file 3dFWHMx.1D.png
----------

I really need your help! (Unfortunately, I couldn't see my message after sending in the google group so I'm sending the same question in email...!) Thanks in advance :)

[Tim Coalson]

The cifti files (.dtseries.nii) have had a small amount of smoothing applied (settable parameter in the fMRISurface batch, suggested to set equal to the grayordinates resolution (voxel size, which vertex spacing is approximately matched to)).  I believe the volumetric versions (.nii.gz) have not had any smoothing applied.

The google group is moderated, messages won't always show up immediately (particularly on weekends).  Please try not to send duplicates.

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/CAJXzLRu0cPGO%2BdNwgjZ6C9UP_3Sh1ZtbmLny4q%3Dh7b%3DrEj4UtA%40mail.gmail.com.

[Glasser, Matthew]

No volume smoothing.  Volume-based data is not recommended for primary use.  Also, we have just released new data, which is now recommended for processing that includes improved spatial ICA cleanup and temporal ICA cleanup.

 

${StudyFolder}/${Subject}/MNINonLinear/Results/${fMRIName}/${fMRIName}_Atlas_MSMAll_hp${hp}_clean_rclean_tclean.dtseries.nii.

 

Matt.

To view this discussion visit https://groups.google.com/a/humanconnectome.org/d/msgid/hcp-users/CANm3QU4fNgZnr%3DKC00MmjSwLO1W1c-hFejn80S-RR_LqaNKwew%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.

[Bokyeong Jeong]

I really appreciate your help!
I've never worked with grayordinate data before and have only dealt with volume-based time series data, so I’m wondering if it’s appropriate to apply the 268 Shen Atlas parcellation to the CIFTI-formatted file you recommended.
I'm mainly planning to perform ISC (Inter-Subject Correlation)-based analysis(https://naturalistic-data.org/content/Intersubject_Correlation.html).

Thanks in advance!

2025년 9월 3일 (수) 오전 10:31, 'Glasser, Matthew' via HCP-Users <hcp-...@humanconnectome.org>님이 작성:

To view this discussion visit https://groups.google.com/a/humanconnectome.org/d/msgid/hcp-users/AE4FCA9D-2A04-4AE7-B8FE-AE437734504A%40wustl.edu.

[Glasser, Matthew]

Not sure about that atlas, but we recommend the HCP’s multi-modal atlas of human cortical areas: https://balsa.wustl.edu/file/87B9N

 

That is read made for HCP data.

Matt.

To view this discussion visit https://groups.google.com/a/humanconnectome.org/d/msgid/hcp-users/CAJXzLRvN7P7w6rGQgnhw8VhvC8jVv%2Bj5yCquONn%3Dgm26t1srjw%40mail.gmail.com.

[Bokyeong Jeong]

The mentioned atlas seems to be used for parcellating cortical areas.

Is it possible to examine both the cortex and subcortical regions using a CIFTI file?

Thank you for your help!

2025년 9월 3일 (수) 오전 10:52, 'Glasser, Matthew' via HCP-Users <hcp-...@humanconnectome.org>님이 작성:

To view this discussion visit https://groups.google.com/a/humanconnectome.org/d/msgid/hcp-users/6ACE16F0-D62B-4DE3-8844-B5FDC25FD93E%40wustl.edu.

[Glasser, Matthew]

The one I gave has major subcortical structures in it.  A multi-modal parcellation of non-neocortical areas is in progress.

To view this discussion visit https://groups.google.com/a/humanconnectome.org/d/msgid/hcp-users/CAJXzLRuzzFc8UO%3DCg%2B3ix3tPUZqfDBt02h4y_cW8ZkVARVP_hA%40mail.gmail.com.