It’s also worth nothing that if you have run from the GUI, the outputs will be in a resampled MNI space ( default 4mm ) which FSLview cannot display simultaneously with a 2mm MNI image. Our great new ( in FSL 5.0.10 ) visualisation tool FSLeyes does not have this restriction - I strongly recommend you try it out.

Matthew

--------------------------------

Dr Matthew Webster

FMRIB Centre

John Radcliffe Hospital

University of Oxford

On 8 Sep 2017, at 16:15, Janine Bijsterbosch <[log in to unmask]> wrote:

Hi Albert,

Melodic (run from the command line) does not apply any transforms for you, so if you are using the concat option for a group ICA analysis you should make sure that the input data from all 22 subjects has been transformed into MNI space (for example using applywarp). You can find more information about this here: http://fsl.fmrib.ox.ac.uk/fslcourse/lectures/practicals/ica/index.html#scripting

Best wishes,

Janine

On 8 Sep 2017, at 15:55, Albert <[log in to unmask]> wrote:

Dear FSL experts,

I have run MELODIC analysis with 22 subjects, using as reference atlas MNI150 brain with 2 mm of resolution.

In the subjects' folders, I see all the outputs and when I open them with FSLview I see also that they have MNI coordinates; so, all the computation seem to be correct.

But, when I try to overimposed one atlas, e.g. Harvard Oxford with 2 mm, I receive a message that "ALL OVERLAYS MUST HAVE SAME DIMENSIONS AS THE BASE IMAGE!" ... so this is strange because they should have the same dimensions ...

Thanks for any hints

Best
Albert

-----
Dr Janine Bijsterbosch
Postdoctoral Researcher
FMRIB Centre, University of Oxford
John Radcliffe Hospital
Oxford, United Kingdom
[log in to unmask]