Dear Bian,
There is no way to extract from SPM solution the activity for each
voxel in the brain. What is stored in D.inv{D.val}.inverse.J are the
waveforms for all the vertices in the cortical mesh. However, the way
we recommend using this information is by generating images for the
time-frequency window of interest and subjecting these images to
statistical analysis. One of the reasons for this is that vertices on
the mesh account for some sensor activity largely because of their
orientation and thus activations may be distributed between vertices
in not very meaningful way. This is fixed when the images are
generated by computing the power and smoothing first on the mesh and
them in 3D. If you want to extract source waveforms based on a source
model with a small number of sources you can use the 'Extract dipole
waveforms' tool included in MEEGTools toolbox
(spm_eeg_dipole_waveforms).
Best,
Vladimir
On Tue, Oct 6, 2009 at 7:27 PM, li bian <[log in to unmask]> wrote:
> Dear all,
>
> I have some questions regarding the EEG source localization with SPM8. Do
> you know how the source localization data come out in SPM? Does it store a
> value in a physical location? If so, can we extract the source data for each
> brain voxel from the Matlab?
>
> Thanks and best regards,
>
> Bian
>
> ________________________________
> Windows Live Hotmail: Your friends can get your Facebook updates, right from
> HotmailŪ.
|