Multivariatetemplateconstruction results

[A Shima]

Hello,

This is the first time that I ask help in this forum, and I want some help for using ANTS.

I have already done Multivariatetemplateconstruction2.sh to create the templates from T1 images, and got the files named;

subject_InverseWarp.nii.gz
subject_Warp.nii.gz
subject_GenericAffine.mat
subject_WarpedToTemplate.nii.gz
subject*_Repaired.nii.gz

for each subjects and the created template.

I have just checked whether the each subject'ssubject_WarpedToTemplate.nii.gz is the same as the results by applying the subject_GenericAffine.mat and subject*_Warp.nii.gz for the original T1 images by

antsApplyTransforms -d 3 -e 0 -i subject_T1nii -r created_template0.nii.gz -o Warped_subject_T1.nii.gz -t [subject*_GenericAffine.mat,0] -t subject*_Warp.nii.gz

However, the results of the subject_WarpedToTemplate.nii.gz and Warped_subject_T1.nii.gz were not the same, and the results are consistent for the each subjects.

Am I correct that the results subject_WarpedToTemplate.nii.gz and Warped_subject_T1.nii.gz have to be the same, and the usage of the antsApplyTransforms?

I appreciate any help.

[C. Rossmanith]

Hi,

I think you should switch the order of the two transforms: the affine transform should be applied first what means that it has to appear last in the command line.

Christina

[A Shima]

Christina,

I appreciate your reply.
Switching the order of the two transforms calculated correct transformation,as you noted. However the intensity is very slightly different. Are there any possibilities for the change of the intensity? (The difference between subject_WarpedToTemplate.nii.gz and Warped_subject_T1.nii.gz, as I defined above.)

I am sorry to bother you many times.

Atsushi

[Christina Roßmanith]

Multivariatetemplateconstruction2.sh does bias field correction unless you switch it off. Maybe this explains the difference in intensities?