On Wed, 12 Jan 2005, Audrey Duarte wrote:
> Hello, I have been using the unwrap option in the volumes toolbox in SPM2
> to correct aliasing artifacts in ASLpMRI images. This works very nicely but
> the newly written images have a different intensity scale than they did
> previously (e.g .05 not 1). As far as I can tell, the actual intensity
> values in specific voxels have not changed, however. Does anyone have
> experience with this? Thanks, Audrey
This may happen because spm_write_vol (the SPM function that writes out
the unwrapped image volume) rescales your data so that they use the full
dynamic range of the image datatype. So, if your image has voxel values
ranging from 0 to 4096, but your datatype supports a range from 0 to
65536, then 4096 will be mapped to 65536 and a scaling factor of
1/16=0.0625 will be used.
This scaling factor is an SPM specific thing, other software will likely
get into trouble (i.e. display wrong voxel intensities) when reading SPM
scaled analyze files. If you need to read in a time series of SPM images
into another software, then you need to rescale the time series to a
common scaling factor, otherwise you will get wrong results. I have
implemented this in a new version of the volumes toolbox, which I will
send to you off list, since my download page is not working at present.
Hope this helps,
--
Volkmar Glauche
-
Department of Neurology E-Mail [log in to unmask]
UKE Hamburg Phone 49(0)40-42803-5781
Martinistr. 52 Fax 49(0)40-42803-9955
20246 Hamburg
|