 |
 |
Dear Pegah, When you coregister in prepare it doesn't keep your head model. It's just done to transform the electrode locations so that it'll make sense to transform them to 2D. You should coregister in either 3D GUI or in batch. If you have the same head model for all subjects you can copy it by copying the whole inv structure, like inv = D1.inv; D2.inv = inv; save(D2); I think the fact that you just copied the mesh created some inconsistency. If you used DCM with ECD I wouldn't worry at all about the mesh problem. If you used imaging, it's also probably OK if there was no error. Vladimir On Thu, Aug 23, 2012 at 1:06 PM, Tayaranian Hosseini P. <[log in to unmask]> wrote: > Dear all, > > > > I have 15 datasets and I have coregistered them all in the main GUI of SPM > EEG using the Prepare SPM option. Then I wrote a script to apply a specific > DCM to all of them to run it overnight but as soon as the first dataset was > loaded, it started asking for the name of the fiducials in the same window > that GUI calls (Select/Type/Click…). I found out that it is trying to fill > D.inv.forward.mesh parameter. I have one channel-location file for all > subjects and coregister it to template MRI. So, I coregistered my first > dataset again, saved the D.inv.forward.mesh structure and copied it to all > my next datasets. So, it did not ask for fiducials anymore. > > > > After DCM analyses finished I tried to apply source reconstruction on the > same datasets and because the forward mesh was already there, I just needed > to press “invert” button but I faced a problem in spdiag function. There is > an “Exceed Matrix Dimension” error. The saved mesh has 5124 voxels but the > new one tries 8196. So, I tried template=>coregister=>forward and then > invert and it worked this time. > > > > 1. Now I am not sure why the two coregistrations are different and > also why it asked for coregistration in DCM in first place when I had > already done it using Prepare function. > > > > 2. Should I now be concerned about the results of DCM on my data? > Should I trust them or should I apply DCM again after source reconstruction? > > > > > > Best, > > Pegah > > -------------------------------------------------------- > > Pegah Tayaranian Hosseini > > PhD Student > > Room 4077, Tizard building (13) > > Institute of Sound and Vibration Research > > University of Southampton, SO17 1BJ, UK > > > > Tel: 023 8059 2850 > > email: [log in to unmask] > >