On 27 June 2013 15:09, Lückmann, Helen (Stud. FPN) <[log in to unmask]> wrote:

Hi Tudor,

sorry, I didn't mean to say that it has to be the pathways. I was just implying that the error messages (both) can occur due to wrong pathways and that nothing substantial has to be wrong with your data ;)
But if all your pathways are given correctly in both cases (i.e., starting from the respective current directories), then I cannot provide any more help, sorry.

Cheers,
Helen

________________________________________
From: FSL - FMRIB's Software Library [[log in to unmask]] On Behalf Of Tudor Popescu [[log in to unmask]]

Sent: 27 June 2013 13:27

To: [log in to unmask]
Subject: Re: [FSL] randomise: "ERROR: Could not open image"

Hi Helen

Thanks for suggesting this, but unfortunately I don't think this is the problem in my case, as I have no 'dodgy' characters in the path to my VBM directory or in my file names, and my call to randomise did not include any other directory/file names. Where exactly did you say you removed a backslah char from?

Thanks again,
Tudor

On 26 June 2013 22:19, Lückmann, Helen (Stud. FPN) <[log in to unmask]<mailto:[log in to unmask]>> wrote:
Hi Tudor,

I got the same error message today and it may be merely related to an error in the directory that you are providing for the file. At least, it disappeared after I removed an error from my file-directory (a / in the beginning).
Cheers,
Helen
________________________________________

From: FSL - FMRIB's Software Library [[log in to unmask]<mailto:[log in to unmask]>] On Behalf Of Tudor Popescu [[log in to unmask]<mailto:[log in to unmask]>]

Sent: 26 June 2013 23:03

To: [log in to unmask]<mailto:[log in to unmask]>

Subject: Re: [FSL] randomise: "ERROR: Could not open image"

Hi Rosalia

Thanks for your reply. The input to randomise does indeed need to be a 4D pre-processed image, which is what my image (defined, for each subject, as the difference between two scans) is. The 4D image contains the pre-processed image of each subject (with the difference taken for each)

Tudor

On 26 June 2013 19:00, Rosalia Dacosta Aguayo <[log in to unmask]<mailto:[log in to unmask]><mailto:[log in to unmask]<mailto:[log in to unmask]>>> wrote:

Hi Tudor,

Not sure, but I think randomise needs 4D file (containing all preprocessed subject files )as input for working. Is this the case for your all diff strut...file?

Best regards,
Rosalia.

El 26/06/2013 18:40, "Tudor Popescu" <[log in to unmask]<mailto:[log in to unmask]><mailto:[log in to unmask]<mailto:[log in to unmask]>>> escribió:

Dear experts,

I ran the following randomise command, from my /VBM/stats/ folder, after having done all the VBM pre-processing:

randomise -i all_diff_struc_GM_to_template_GM_mod.nii.gz -m GM_mask -o fslvbm -d design.mat -t design.con -T -n 5000

The file all_diff_struc_GM_to_template_GM_mod.nii.gz has been calculated with fslmaths and looks fine when viewed in fslview's Movie Mode. Still, I get the error:

ERROR: Could not open image all_diff_struc_GM_to_template_GM_modTrace: read_volume4DROI.

Although I know randomise needs to be run from /stats/, I tried running the same command from one folder level up (/VBM/), but then it says:

ERROR: Could not open image GM_mask

Can anyone suggest what could be wrong? Thanks!

Tudor