once every time you click "dilate label" On Sat, 7 Nov 2009, Martin Kavec wrote:

Hi Bruce,

if I dilate an arbitrarily shaped label in tksurfer, how much is it actually dilated?

Thanks, Martin

On Sat, Nov 7, 2009 at 8:18 PM, Bruce Fischl fischl@nmr.mgh.harvard.eduwrote:

...

Hi Yigal,

yes, you can select the vertex, make a label out of it, then dilate it however many times you want. This will be approximately spherical. Alternatively you could write some code to use the ?h.sphere surface to compute geodesic distances (great circles) to do it more exactly, but we don't have anything that will do that (easy enough to do in matlab though).

cheers Bruce

On Sat, 7 Nov 2009, Yigal Agam wrote:

...

Hi,

In surface-based analysis, is it possible to create an ROI label that is a circle of a certain radius around a given vertex?

Thanks, Yigal _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer

---

Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer

Hi Martin,

it will dilate by one neighbor in the mess, regardless of distance. The neighbors start out about 1mm apart, but there is some variation (which is why it won't end up being exactly circular).

cheers, Bruce On Sat, 7 Nov 2009, Martin Kavec wrote:

:) thanks. In volume images dilation is usually by 1 voxel, but as far as I understand, label is a surface represented by a mesh where a distance between vertices does not need to be the same. So if I want to dilate the label by 10 mm do I click "dilate label" 10x.

Thanks,

Martin

On Sat, Nov 7, 2009 at 8:36 PM, Bruce Fischl fischl@nmr.mgh.harvard.eduwrote:

...

once every time you click "dilate label"

On Sat, 7 Nov 2009, Martin Kavec wrote:

Hi Bruce,

...

if I dilate an arbitrarily shaped label in tksurfer, how much is it actually dilated?

Thanks, Martin

On Sat, Nov 7, 2009 at 8:18 PM, Bruce Fischl <fischl@nmr.mgh.harvard.edu

...

wrote:

Hi Yigal,

...

yes, you can select the vertex, make a label out of it, then dilate it however many times you want. This will be approximately spherical. Alternatively you could write some code to use the ?h.sphere surface to compute geodesic distances (great circles) to do it more exactly, but we don't have anything that will do that (easy enough to do in matlab though).

cheers Bruce

On Sat, 7 Nov 2009, Yigal Agam wrote:

Hi,

...

In surface-based analysis, is it possible to create an ROI label that is a circle of a certain radius around a given vertex?

Thanks, Yigal _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer

---

Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer

Hi Rudolph,

and thanks for the offer. I think I have to appologize to the author of the thread for jumping into his discussion. I just felt that the topic of my question was so close to his that I didn't resist.

In fact in my analysis, label dilation as implemented in tksurfer and suggested by Bruce is sufficient for my purpose.

Thanks again,

Martin

On Sunday 08 November 2009 23:37:18 Rudolph Pienaar wrote:

Hi Martin --

There is an app in the development tree, 'mris_pmake' that can be used to generate fixed-radius regions about arbitrary vertices. It's scriptable, so you can feed it a list of vertex indices to build circular ROIs about.

It should be available in the 5.0 release of FreeSurfer, but if you'd like to use it already, send me an email and I can send you a beta version.

Cheers -=R

Martin Kavec wrote:

...

:)

thanks. In volume images dilation is usually by 1 voxel, but as far as I understand, label is a surface represented by a mesh where a distance between vertices does not need to be the same. So if I want to dilate the label by 10 mm do I click "dilate label" 10x.

Thanks,

Martin