Converting BRIK to dset error

Hi,

I am writing to get some help with converting the BRIK file to dset file.

I would like to open GLM results (originally made as Nifti and it was converted to BRIK file) on SUMA to check whether the results are overlapped with ROI masks drawn by SUMA.

By using 3dcopy, the Nifti file was converted to BRIK file successfully. Furthermore, it can be opened well on AFNI and Suma also.

However, I would like to check whether it overlaps with the ROI masks. Hence, I need to convert the BRIK file to dset by following AFNI guidelines on youtube ( [AFNI Academy] View overlapping surface ROIs using SUMA GUI (youtube.com))

Hence, I tried to convert the BRIK file to dset by using the ConverDset function of the AFNI. Unfortunately, the dset file does not show any results on the SUMA.

Although the ConvertDset function creates dset file, there are error messages. It will be attached below. Furthermore, the code for ConvertDset function is also attached.

Could you let me know what I missed?

Error SUMA_Dset_Type (suma_datasets.c:4915):
Null Name
Error SUMA_Dset_Type (suma_datasets.c:4915):
Null Name
Error SUMA_Dset_Type (suma_datasets.c:4915):
Null Name
Error SUMA_Dset_Type (suma_datasets.c:4915):
Null Name
Error SUMA_Dset_Type (suma_datasets.c:4915):
Null Name
Error SUMA_Dset_Type (suma_datasets.c:4915):
Null Name
Error SUMA_Dset_Type (suma_datasets.c:4915):
Null Name
Error SUMA_Dset_Type (suma_datasets.c:4915):
Null Name
Error SUMA_Dset_Type (suma_datasets.c:4915):
Null Name

AFNI version info (afni -ver): Jan 31 2024 (Version AFNI_24.0.04 'Caracalla')

code text  # or delete if not needed

ConvertDset \
    -o_niml \
    -input uncorrected+tlrc.BRIK.gz\
    -prefix uncorrected

Howdy-

So, if I understand correctly, you have volumetric information (NIFTI or BRIK/HEAD file, either is fine in AFNI), and you have a surface ROI, and you want to see if they overlap? To do so, you want to project the volumetric dataset onto an existing SUMA surface (presumably created with FreeSurfer's recon-all and AFNI's @SUMA_Make_Spec_FS).

Note that you could probably use the afni and suma GUIs together with talking to check this out:
[AFNI Academy] SUMA introduction: Background and start GUI.

But to project the volumetric data onto the SUMA surface, you can use 3dVol2Surf. Examples are shown in the help file, such as whether you want to use 1 or 2 surfaces for the projection, but one might be:

    1. Apply a single surface mask to output volume values over
       each surface node.  Output is one value per sub-brick
       (per surface node).

    3dVol2Surf                                \
       -spec         fred.spec                \
       -surf_A       smoothwm                 \
       -sv           fred_anat+orig           \
       -grid_parent  fred_anat+orig           \
       -map_func     mask                     \
       -out_1D       fred_anat_vals.1D

... and instead of -out_1D .. you could perhaps use: -out_niml fred_anat_vals.niml.dset for NIML format. That should give you a surface-based projection of your ROI, which is input via -grid_parent ...

--pt

Thank you so much. It helps to transform ROI masks.

I have one more question.

Is there a smoothwm file has both hemisphere on MNI file? Because the GLM results file has both sides of datasets and is aligned on MNI space, I think both sides of smooth files utilize the 3dVol2Surf function. However, I only can see separated left and right hemispheres of smoothwm file on the MNI file of the AFNI.

Hi-

After running @SUMA_Make_Spec_FS on the FreeSurfer results, you should see multiple std.*both.spec files---the different numbers just refer to different grid/mesh resolutions (larger number is finer mesh). You should be able to enter that for your -spec .. file input, to have both hemispheres.

--pt

Thank you so much!