Dear Daniela,

You should have an image that looks like a standard brain
image (the magnitude image) and another image that
looks like a phase image of some kind.  It is probably a phase
difference image if you only have one of them.  So if your
INT16 image looks like a phase difference image then you
need to scale it first to be in units of radians.  This is probably
done by diving by 2048 and multiplying by pi (as in step 2b,
but only for your single image).  However, you will need to
check with your local scanner operator and check the values
in the image to see if that makes sense.  Unfortunately different
types of scanners produce different output so I cannot be more
specific here.

For the magnitude image it is *much* better to use the
magnitude image that comes with the fieldmap.  Every
fieldmap sequence should produce one and you should
be able to access it.  If you really cannot get this (say,
because it was not saved in the session) then it is possible
to use the T1 structural but it is *very* difficult to do so, as
it is often very difficult to register it to the phase difference
image, which is what you must do.  Typically it requires you
to manually make a brain mask from the phase difference
image and register that to a brain mask from the T1 structural,
but even with these it often requires a lot of intervention.  So
really try to find your fieldmap magnitude image first.

I hope this helps.
All the best,
	Mark


On 20 Oct 2010, at 08:02, Daniela Gebauer wrote:

> Dear FSL-experts,
>
> I try to make fieldmap images for FEAT but am not quite sure about  
> the Processing Steps:
>
> My datatype info is:
>> fslinfo FM_BeutleCa.nii
> .gz
> data_type INT16
> dim1 64
> dim2 64
> dim3 36
> dim4 1
> datatype 4
> pixdim1 3.0000000000
> pixdim2 3.0000000000
> pixdim3 3.7499995232
> pixdim4 0.4000000060
> cal_max 0.0000
> cal_min 0.0000
> file_type NIFTI-1+
>
> I read, that INT16 is a float field map (scaled unwrapped phase  
> difference img) and it is unmasked as a physician told me.
>
> So do I continue with the steps 1b, 1c, 4b and 5? (for a single,  
> real fieldmap?) or are there any other steps beforehand?
>
> ad Step 1b: Can I use the T1 structural scan as magnitude image?  
> (undistorted and registered) - and if yes:
> do I take the BET-image?
>
> My BET-T1 is also having different dim and pixdim / resolution - so  
> what to do?
>
> data_type
> dim1           176
> dim2           256
> dim3           166
> dim4           1
> datatype       4
> pixdim1        1.0000000000
> pixdim2        1.0000000000
> pixdim3        1.0000000000
> pixdim4        0.0000000000
> cal_max        93.0000
> cal_min        0.0000
> glmax          93
> glmin          0
> origin1        0
> origin2        0
> origin3        0
> file_type      ANALYZE-7.5
>
>
> I would be really grateful for your help, thank you very much in  
> advance, Daniela
>