[caret-users] misaligned metric and structural file

Donna Hanlon donna at brainvis.wustl.edu
Thu Jul 29 11:37:44 CDT 2004

Hi Sophie,

Some things to check:

1.  The anatomical to which the EPI scans were aligned is in the same 
voxdims and orientation as the uncropped anatomical (*.LR.*mnc) used to 
generate the SureFit surface.
2. The fiducial surface you're using is the one generated by SureFit, 
with no transforms applied.
3.  You loaded the appropriate *cropped* params file, which tells the 
mapper the Xmin, Ymin, and Zmin values by which the surface must be 
translated, along with any padding parameters, if this was not a full 

What I confess is not clear to me is if the map_fmri_to_surface program 
will try to center the volume based on the SPM originator voxel in the 
.hdr, when mapping to an individual surface.  John Harwell would know. 
 In that case, then the vtk surface would need to be translated -- 
probably by this amount:

-78,-112,-50 (default bounding box)
-90,-126,-72 (template bounding box)

This assumes cubic 1mm voxels, as required by SureFit.


On 07/29/2004 11:30 AM, Sophie Riches wrote:

> Hello again! Sorry to keep bothering you but i keep getting foiled!
> I have aligned a set of EPI fMRI scans to a structural T1 scan in SPM 
> 99 before processing the fMRI data.
> The output from SPM looks okay - the activation is where I expected so 
> I think the orientation is ok and in normal SPM orientation. When 
> overlaid on slices it appears to be nicely aligned.
> However when I map the fMRI to the reconsrtucted surface (from 
> SureFit) it is quite clearly not aligned at all. I am using the 
> map_frmi program on windows and have put in the fidiucial file as the 
> mapping surface,
> Have I missed out a step? I have used the parameter file for the 
> surface and it has set the origin automatically .(although I thought 
> the origin of a surface made in SureFit would be 0,0,0 so I tried that 
> and still get it misaligned.) It certainly looks as though it is in 
> the right orientation but is not centred on the same origin.
> Any ideas?
> Sophei
> ------------------------------------------------------------------------
> *ALL-NEW Yahoo! Messenger* 
> <http://uk.rd.yahoo.com/evt=21626/*http://uk.messenger.yahoo.com>* - 
> all new features - even more fun!** *
>caret-users mailing list
>caret-users at brainvis.wustl.edu

More information about the caret-users mailing list