I'm not sure which version of SPM you are using. If you are running SPM96,
then you will also need to convert the data in the templates, canonical and
apriori directories to Linux format, as coregister requires these images
during the processing.
If you are running SPM99b, then the byte conversion should not really be
necessary. If this is the case, I don't know what the source of the problem
is, but it would be slightly easier to diagnose if you send the whole error
report.
Good luck,
-John
> I am just beginning to learn SPM and started from analyzing the data in the
> SPM website(i.e., MoAEpilot data, for fMRI). I have a trouble in
> Coregistration step.
>
> First, based on suggestions in the SPM mailbase, I realigned the fM* images
> specifying options as Coregister&reslice => sinc interpolation => mean image
> only. It was successful.
>
> Next, I tried to coregister sM* structural image, by options as Coregister
> only => target=sM*img(T1) => object=mean*.img(T2). Then SPM stopped, warning
> "unrecognised datatype".
>
> I use Linux based PC environment so that I converted all image files into
> Linux format, by Matthew Brett's byte_conversion tool. I am sure this
> small-to-big endian conversion was successful; they all look beautiful by
> DISPLAY command.
>
> I may misunderstand the coregister procedure. Could anyone provide any
> suggestions?
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|