how best to use ANTS with brains with lesions

[satra]

is there an example of running antsRegistration with a moving image that has a lesion to a fixed image that is an intact brain? we do have a binary mask of the lesion and both images are brain only. ping: @mgeddes

[stnava]

just use

ImageMath 3 non_lesion.nii.gz Neg lesion.nii.gz

ImageMath 3 non_lesion.nii.gz ME non_lesion.nii.gz 2

standard antsRegistration call but add

-x non_lesion.nii.gz

also --- use the moving image as the fixed image .... then apply inverse
mappings

brian

On Fri, Oct 18, 2013 at 2:09 PM, Satrajit Ghosh notifications@github.comwrote:

is there an example of running antsRegistration with a moving image that
has a lesion to a fixed image that is an intact brain? we do have a binary
mask of the lesion and both images are brain only.

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/48
.

[satra]

@stnava - thanks - but can -x take only mask? according to the help it needs two.

also --- use the moving image as the fixed image .... then apply inverse mappings

just to clarify, are you saying i should register the intact brain to the lesioned brain as opposed to the other way around?

[stnava]

Yes to both questions
On Oct 18, 2013 4:58 PM, "Satrajit Ghosh" notifications@github.com wrote:

@stnava https://github.com/stnava - thanks - but can -x take only mask?
according to the help it needs two.

also --- use the moving image as the fixed image .... then apply inverse
mappings

just to clarify, are you saying i should register the intact brain to the
lesioned brain as opposed to the other way around?

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/48#issuecomment-26629180
.

[satra]

thanks brian, this worked reasonably well. closing.