Dear Spm'rs
I have been looking at the use of SPM5 for pooling T1 scans from
different scanners. If I employ SPM5 segmentation with its default
settings for unified normalisation and classification I find that the
resultant segmentations are strongly influenced by scanner noise.
Outwith SPM5's unified method I can limit the effects of scanner
noise. I was wondering if within the unified method there were means
that would help compensate for differences in short form noise (ie
grain in the T1 profiles).
Bill Moorhead
Psychiatry
Edinburgh
--
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.
|