> Recently I have had to adjust the orgins of the
> images. I find the AC-PC, enter the new coordinates
> into the boxes below, reorient the images, and then
> check them-- the images look good. All the origins for
> the T1 template, anatomical, and functional images match.
> There is an old HDR file and then there is an updated Microsoft access
> file linking the new coordinates (I believe).
That file should be a Matlab .mat file.
>
> I then cogregister the images (no reslice) and then
> normalize. When I normalize the origins appear to go
> back to the old incorrect orgins! In fact, the normalized
> images look really weird and if one tries to analyze them
> the activation for primary somatosensory cortex gets localized
> in the temporal lobe!
Unless there has been a problem with the coregistration, it sounds like the spatial normalisation is
going wrong, which can happen for a number of reasons:
* Limited FOV in the image
* Strong intensity nonuniformities or other artifacts
* Pathology
* Different contrast between the image and template that you match it to
* File format problems
- wrong voxel sizes (should be in millimeters)
- intensity scaling problems (check range of image intensities)
* Initial poor starting estimates (although I know that your origin should be OK)
Check the above and see if any of them could be the cause of the problem.
Best regards,
-John
--
Dr John Ashburner.
Functional Imaging Lab., 12 Queen Square, London WC1N 3BG, UK.
tel: +44 (0)20 78337491 or +44 (0)20 78373611 x4381
fax: +44 (0)20 78131420 http://www.fil.ion.ucl.ac.uk/~john
|