Good morning,
I wrote some time ago about my registration-problems occurring only in
"certain" slices. I think I may have found somekind of a solution, but I
would like to get some justification for it...
To guide flirt to "stretch" also the problem slices to the whole area of the
reference slice, I used as -init mat-file one of the upper slice output
mat-files. This way flirt managed to produce quite nice registration. Is this
a proper way to use -init? The result varies slightly, if I use mat from 1st,
2nd or 3rd slice.
With best regards,
Pasi Tuunanen
|