Many thanks, Peter. Looks really nice. I shall implement.
Hi Peter–
Just got through trying out your approach and it looks really nice. Everything lines up. One remaining issue, though, is that some blurring gets introduced into the -input image during alignment to the -base. Is there a way to avoid this?
Many thanks…
Paul
Hi Paul,
You can change the -master to be the same as the input, that should avoid the system trying to add slices/interpolate.
-PM
Thanks much, Peter. Traveling now but will try once I’m back at my desk…
Hey Peter–
Still looking interpolation-y.
I ran:
3dAllineate -base anat_ns.nii -input t1_dyn_contrast_volreg_ave_zp_ns.nii -master t1_dyn_contrast_volreg_ave_zp_ns.nii \
-prefix t1_dyn_contrast_volreg_ave_zp_ns_al.nii -1Dmatrix_save t1_dyn_2_spgr.1D -cost mi -final wsinc5 -source_automask \
-autoweight -twopass -warp shift_rotate -verb -overwrite
where:
anat_ns.nii = high res whole-brain anatomical without skull
t1_dyn_contrast_volreg_ave_zp_ns.nii = low res T1, zeropadded, sans skull
Results represented in attached picture. You can see that the resolution has, indeed, been defined by the low-res T1 but we still get lots of interpolation (e.g., where the zero-pad area is not zero).
Advice appreciated. :-S
