DICOM-fields based registration can be invaluable as the start point for
further FLIRTing in the cases like mine. That is strange that it doesn't
get used.
And indeed we should start collecting field maps for our bolds ;-)
On Tue, 10 Apr 2007, Steve Smith wrote:
> Hi,
> In theory you could try to use the DICOM headers instead of registration but AFAIK no software does
> this, partly as DICOM is such a pain to work with. Also, as you have distortion in your BOLD data
> that is not present in the structural, you are likely to end up with better registrations by using
> (e.g.) PRELUDE+FUGUE unwarping (using a fieldmap) and then FLIRT registration. All of this can
> easily be done with FEAT if you have the fieldmap.
> Cheers, Steve.
--
.-.
=------------------------------ /v\ ----------------------------=
Keep in touch // \\ ([log in to unmask])onerussian.com
Yaroslav Halchenko /( )\ ICQ#: 60653192
Linux User ^^-^^ [175555]
|