Hi Brian, I read in a surface file and it spit out 3 coordinates, do you know if they are RAS and also there is no vertex number with them, so does the vertex number start at zero and count up then?

they are RAS coords. surface vertices go from 0-> nverts matrix(in matlab) will go from 1-> nverts+1

One more quick question. Do I have to read in the hemisphere's separately or is there a surface file that has both of them together (either the orig surface or the smoothwm surface)?

each hemisphere separately

Thanks so much for you help! Sarah

thank you, brian

Brian T. Quinn wrote:

...

hi Sarah-

you can you the matlab code in $FREESURFER/matlab to do this. use read_surf.m for this.

-- brian

On Fri, 15 Jul 2005, Sarah Rugheimer wrote:

...

Hello,

Does anybody know if there is a file with or a way to obtain a mapping of the vertex number with say the RAS coordinate system. I know when you are in the tksurfer window by using the mouse/cursor you can see each vertex number with the corresponding RAS coordinate, but is there a way to output all the vertex numbers with the cooresponding RAS coordinates to a file or matrix or some such thing.

Thank you very much for any help.

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

Sarah:

Freesurfer surface file formats are documented at:

http://grahamwideman.com/gw/brain/fs/surfacefileformats.htm

... including the "W" file format (scroll down).

The coordinate system for FreeSurfer's vertices is RAS, described in more detail at:

http://grahamwideman.com/gw/brain/fs/coords/fscoords.htm

As for:

My main question is this: Do you know how I can convert the vertex coordinates that I have from my MEG data to the vertex number that freesurfer uses. Are the vertex numbers part of a coordinate system that I can then transform between the two systems?

Presumably your MEG data provides 3D coordinates in some XYZ space defined by your MEG equipment or software. The transformation between that coordinate system and FS's would need to be known to convert vertex *positions* in one space to the other.

In situations where your two types of equipment don't tell you how to ttransform between the two systems (eg: data gathered in entirely different equipment at different times), then you have to perform some sort of registration based on the actual images themselves (or possibly surfaces if that's what you have). If that's your situation then it would certainly be helpful to find someone in your exact same situation in case that might save you from devising it yourself.

FS's vertex numbers themselves have meaning only as IDs within the particular subject -- there's no correspondence even across subjects, unless you perform the intersubject surface registration, which results in a cross-reference. (Then the vertex IDs are still subject-specific, but you at least have a cross-reference.)

(The above of course goes out the window if, while I was not paying attention, FS has sprouted its own MEG feature that I don't know about!).

Hope that helps,

Graham