Dear Ping,
Thank you for your reply, but my template is a betted image. Is the input of fnirt also should be an image non-betted?
if your template is skull-stripped, and you are unable to get a version with the skull intact, it might be that your only choice is to use a betted image.
The problem with that is that the accuracy of your registration will be completely dependent on how well bet has worked. Any small part of the brain that has been removed by bet will be stretched to fill out the void, and any non-brain tissue that has been left will be squashed into the brain. I would strongly urge you to visually inspect all your brain extractions and manually correct any problems before running fnirt on them.
Also, as I said in my previous email, it is not advised to run fnirt without a configuration file.
And I don't want to get the deformation field from the template to the subject space , I want to get the deformation field from the template to the linearly registered image(flirt_T1).So, I tried like this after fnirt:
fnirtfileutils --in=warp_coef.nii.gz --ref=Colin27_Template.nii.gz --out=fnirtfileutils_field
invwarp --ref=flirt_T1.nii.gz --warp=fnirtfileutils_field.nii.gz --out=invfileutils_mni2flirt
I want to know if the 'invfileutils_mni2flirt' file I got is the deformation field of the image I want from the template to the linear registered image(flirt_T1)?
Please see my reply to a later email.
Jesper
ping
Jesper Andersson <[log in to unmask]<mailto:[log in to unmask]>> 于2019年6月20日周四 下午5:01写道：
Dear Ping,
the warp field that takes the template into subject space should be the same size as the subject image (T1 image of the original space), so that is absolutely correct.
But there are a couple of other problems with your commands below
I want to get the warp_field from the standard space template to the linearly registered T1 image. How should I perform invwarp?
I have tried to do this as follows，
"flirt -ref Template -in struc_brain.nii.gz -omat flirt_transf.mat -out flirt_brain
fnirt --in=struc_brain.nii.gz --aff=flirt_transf.mat --cout=fnirt_transfcoef --ref=Template
When you run fnirt you should _not_ use the brain extracted (betted) data. You should use the non-betted (original) images and warp them to the non-betted template (typically the MNI152_T1_2mm.nii.gz image)
You should also use a suitable configuration file. For T1-weighted human images we recommend the T1_2_MNI152_2mm.cnf file. If using it for other types of data that file still provides a good starting point for defining ones own configuration file.
Jesper
applywarp --ref=Template --in=struc_brain.nii.gz --warp=fnirt_transfcoef --out=fnirt_brain
invwarp --ref=struc_brain.nii.gz --warp=fnirt_transfcoef.nii.gz --out=warpfield_into_struc_ space"
But the resulting warp_field is the same size as the T1 image of the original space. This should not be the warp_field from the template to the linear registration image I want to get. So, what should I do for this?
Your help is greatly appreciated.
ping
________________________________
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
________________________________
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
________________________________
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
########################################################################
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
