Hi,

The first one of the fslstats commands is correct (as it multiplies the mean value of the PVE by the volume).  The second one only prints the volume for the non-zero voxels, which will overestimate the tissue volume.

As for your previous email - the FAST outputs are in the native space, not MNI space.

All the best,
Mark

On 27 Oct 2014, at 14:28, Mahmoud <[log in to unmask]> wrote:

Dear Mark,

Suppose I used faslmaths -roi correctly and have the results in hand as brain_roi.
which on of these codes give me the volume of the brain_roi:

fslstats brain_roi –M –V  | awk ‘{print $1 * $3}’                                                                       

 

or 

fslstats brain_roi –M –V  | awk ‘{print $3}’

and what would be the units for that volume ?

Thank you,
Mahmoud

 


On Sat, Oct 25, 2014 at 1:57 PM, Mark Jenkinson <[log in to unmask]> wrote:
Hi,

I'm not 100% sure what the exact calculation is that you are trying to do, but often asymmetries are calculated by taking the images and registering them to a symmetric standard space template, and then flipping the image in the template space to compare things between left and right in a voxel-wise manner.

If you want to calculate left and right summaries instead, then the image after registration to the template can be easily masked using the -roi option in fslmaths.  For example, with the 1mm MNI templates you would use:
  fslmaths <my_image> -roi 0 91 0 -1 0 -1 0 1 <my_image_right>
  fslmaths <my_image> -roi 91 91 0 -1 0 -1 0 1 <my_image_left>
For other templates or other resolutions you'd need to modify these numbers.

All the best,
        Mark


On 24 Oct 2014, at 19:31, Mahmoud <[log in to unmask]> wrote:

> Dear all,
>
> Can anyone let me know how to compute the brain tissue asymmetry index using FSL ?
> I need to have the separated index for CSF, GM and WM.
>
> Thank you,
> Mahmoud