External Email - Use Caution        

Hi Lilla,

 

  1. Many thanks for the assurance.
  2. My structural data was preprocessed in FS while my functional data was preprocessed in FSL. As far as I can tell, the difference lies in the MNI152 standard brain size:
    1. In FSL, the MNI152 1mm brain’s SForm code is MNI152 and its dimensions are anisotropic at 182x218x182.
    2. In FS, the SForm code for the T1 brain in the cvs_avg35 folder (as well as the output from the registration to the MNI152) is Aligned anatomical and its dimensions are isotropic at 256^3.

I want to use FS as I need a more accurate segmentation of the amygdala. Is there a reason the two MNI152 brains would have different numbers of voxels and is there a way to reconcile this so that my structural and functional have the same voxel dimensions?

 

Many thanks,

Liz

 

From: <freesurfer-bounces@nmr.mgh.harvard.edu> on behalf of "Zollei, Lilla,Ph.D." <LZOLLEI@MGH.HARVARD.EDU>
Reply to: Freesurfer support list <freesurfer@nmr.mgh.harvard.edu>
Date: Thursday, 2 July 2020 at 1:07 am
To: Freesurfer support list <freesurfer@nmr.mgh.harvard.edu>
Subject: Re: [Freesurfer] Bounding box dimensions for mri_cvs_register (MNI space)

 

Hi Liz,

 

(1) Given that the included MNI template lives in the 1mm isotropic space, the output will also live there. Your two commands seem to be the right for to achieve what you intend to do. 

 

(2) I suspect there was a dimensionality difference between your fMRI and sMRI images even before the CVS registration was run. You will need to register those two spaces in order to accurately align them. (I am not exactly sure what steps are involved in your FSL preprocessing)

 

Lilla

From: freesurfer-bounces@nmr.mgh.harvard.edu <freesurfer-bounces@nmr.mgh.harvard.edu> on behalf of Elizabeth Haris <e.haris@unsw.edu.au>
Sent: Wednesday, July 1, 2020 3:20 AM
To: Freesurfer support list <freesurfer@nmr.mgh.harvard.edu>
Subject: [Freesurfer] Bounding box dimensions for mri_cvs_register (MNI space)

 

        External Email - Use Caution        

Hi guys,

 

I have a couple of queries regarding the output dimensions for mri_cvs_register.

 

  1. If I am registering my images to MNI space, can the command give me a 2mm image, or will I only get a 1mm image which I then need to convert into a 2mm? I am currently getting a 1mm image and resampling it to a 2mm image afterwards (for fMRI analysis), using the following commands, and wanted to double check the procedure:
    1. CVS command: mri_cvs_register --mov <subjID> --mni
    2. FSL command:

flirt -in <final_CVSmorphed_tocvs_avg35_inMNI152_norm.nii> \

-ref < final_CVSmorphed_tocvs_avg35_inMNI152_norm.nii > -applyisoxfm 2.0 -nosearch \

-out <subject_MNI_2mm.nii>

 

  1. I also noticed that when I resample the image this way, though its dimensions are 2mm^3 the voxels in the bounding box are different to the voxels in my 2mm^3 functional image preprocessed in FSL (FS: 128^3 vs FSL: 109 91 109).
    1. Is this a problem?
    2. If so, is there a way to rectify this?

 

Many thanks!

Liz