Try using the robustfov command on your images before running bet. If you still do not get good results, you will need to manually tweak the bet settings.

Cheers,

Paul

On 2 November 2016 at 04:30, Isobel French <[log in to unmask]> wrote:

Hi Paul,

Do you know-how I can use fslmaths to correct the results on bet? Some of the images I obtained do not give a whole brain.

4 AM, Isobel French <[log in to unmask]> wrote:

I see. Tqvm.

On Thu, Oct 27, 2016 at 4:42 PM, paul mccarthy <[log in to unmask]> wrote:

Hi Isobel,

Those two commands are identical, just with different file names

Cheers,

Paul

On 27 Oct 2016 03:16, "Isobel French" <[log in to unmask]> wrote:

Dear Paul,

I see. But why does the practical say to use the command fslroi dwidata nodif 0 1 then?

Thanks.

On Wed, Oct 26, 2016 at 6:53 PM, paul mccarthy <[log in to unmask]> wrote:

Hi Isobel,

Are you just trying to extract the first 3D volume of a 4D image? In that case, use fslroi as follows:

fslroi data first_volume 0 1

Cheers,

Paul

On 26 October 2016 at 05:10, Isobel French <[log in to unmask]> wrote:

Hi Paul,

I tried carrying out fslroi on an eddy corrected dti data, and the response below was what I got:

[neuros-iMac-2:YEOH_LE/3775/0] neuro% fslroi data 0 1

Usage: fslroi <input> <output> <xmin> <xsize> <ymin> <ysize> <zmin> <zsize>

       fslroi <input> <output> <tmin> <tsize>

       fslroi <input> <output> <xmin> <xsize> <ymin> <ysize> <zmin> <zsize> <tmin> <tsize>

Note: indexing (in both time and space) starts with 0 not 1! Inputting -1 for a size will set it to the full image extent for that dimension.

[neuros-iMac-2:YEOH_LE/3775/0] neuro% 

What does this mean?

Thanks.

On Thu, Oct 6, 2016 at 1:09 PM, Isobel French <[log in to unmask]> wrote:

Hi Paul,

I did what you suggested and it is working. TQvm indeed!

On Wed, Oct 5, 2016 at 5:03 PM, paul mccarthy <[log in to unmask]> wrote:

Hi Isobel,

It seems like there is something already in the default install location (/usr/local/fsl) - maybe a partial (failed) installation. 

Could you check the contents of /usr/loca/ (with "ls -l") ?  If there is already an FSL installation in /usr/local/fsl, remove it and try installing again.

Cheers,

Paul

On 4 October 2016 at 10:45, Isobel French <[log in to unmask]> wrote:

Hi Paul,

I did what you suggested and this is what I got:

Downloading FSL version 5.0.9 (this may take some time)

Checking download...        

[OK] Download is good.

Password:

Password:

Installing FSL software...

[FAILED] "The contents of /usr/local doesn't look like an FSL installtion - will not install here, aborting install" ; 

[neuros-iMac-2:~/Downloads] neuro% 

What does this mean?

On Fri, Sep 30, 2016 at 7:04 PM, paul mccarthy <[log in to unmask]> wrote:

Hi Isobel,

As a guess, it looks like one of the questions asked by the fslinstaller.py script was given an incorrect answer (e.g. 'yes' where there should have been a directory, or no answer [i.e. just press enter]).

I would recommend that you try to run the installer script again.

Cheers,

Paul

On 30 September 2016 at 04:26, Isobel French <[log in to unmask]> wrote:

Hi,

After running the installer script I ran the command echo $FSLDIR and got the following as attached. What should I do now? 

Tqvm.

[neuros-iMac-2:~] neuro% echo $FSLDIR

yes/fsl

[neuros-iMac-2:~] neuro% 

[neuros-iMac-2:~] neuro% flirt -version

flirt: Command not found.

[neuros-iMac-2:~] neuro%