- The added line is THIS COLOR.
- The deleted line is THIS COLOR.
[[FrontPage]]
http://www.grahamwideman.com/gw/brain/afni/alldocs_afni.htm
Assume we have the following data sets in nifti format.
4 T1W 3Ds(T1-1,T1-2,T1-3,T1-4)
2 T2*W 3Ds(T2-1,T2-2)
Scan order: T1-1 T2-1 T2-2 T1-2 T1-3 T1-4
1.Registration among T1s
T1-1>T1-2<T1-3
<T1-4
2.Registration among T2*s
T2-1>T2-2
3.Registration between T1ave and T2*ave
T1ave<T2ave
AFNI commands: 3dvolreg, 3dAllineate 3dcopy　3dAFNItoNIFTI　(3dAFNItoANALYZE)
Processing steps:
human3d 3d.t1-1 -F<RET>
...
human3d 3d.t2-2 -F<RET>
### Registration among T1s ###
3dvolreg -verbose -dfile param_t1_12 -base 3d.t1-2.nii -prefix 3d.t1-1reg 3d.t1-1.nii
3dvolreg -verbose -dfile param_t1_23 -base 3d.t1-2.nii -prefix 3d.t1-3reg 3d.t1-3.nii
3dvolreg -verbose -dfile param_t1_24 -base 3d.t1-2.nii -prefix 3d.t1-4reg 3d.t1-4.nii
### You should check param files here ###
3dcopy 3d.t1-2.nii 3d.t1-2
cat 3d.t1-1reg+orig.BRIK 3d.t1-2+orig.BRIK 3d.t1-3reg+orig.BRIK 3d.t1-4reg+orig.BRIK > 3d.t1all+orig.BRIK
mstat 3d.t1all+orig.BRIK 4 -ave
cp 3d.t1-1reg+orig.HEAD 3d.t1all+orig-ave.HEAD
### Registration among T2*s ###
3dvolreg -verbose -dfile param_t2_12 -base 3d.t2-2.nii -prefix 3d.t2-1reg 3d.t2-1.nii
### You should check param files here ###
3dcopy 3d.t2-2.nii 3d.t2-2
mapcal 3d.t2-1reg+orig.BRIK ave 3d.t2-2+orig.BRIK 3d.t2ave+orig.BRIK
cp 3d.t2-1reg+orig.HEAD 3d.t2ave+orig.HEAD
### Registration between T1ave and T2*ave
3dAllineate -verb -warp shift_rotate -base 3d.t1ave+orig -source 3d.t2ave+orig -prefix 3d.t2ave_reg
mapcal 3d.t2ave+orig.BRIK lt 0.05 mask.sdt
mapcal 3d.t2ave+orig.BRIK + mask.sdt 3d.t2avethre+orig.sdt
mapcal 3d.t1ave+orig-ave.BRIK / 3d.t2avethre+orig.sdt 3d.t1t2ave+orig.BRIK
cp 3d.t2-2+orig.HEAD 3d.t1t2ave+orig.HEAD
3dAFNItoNIFTI -prefix result 3d.t1t2ave+orig
(3dAFNItoANALYZE -orient LPI result 3d.t1-4reg+orig)
(*)If you find something is missing or too much sandy noise outside of the brain, you need to adjust the threshold value for "mapcal 3d.t2ave+orig.BRIK lt 0.05 mask.sdt
".