Thank you for your prompt reply.
I tried to run these steps individually for 4 subjects and then I tried to
run the group inversion.
It starts well, but then there is an error. Maybe someone has experienced it
before:
Error using file2mat
Cant open file.
Error in file_array/subsref>multifile2mat (line 177)
val(cc(i)+1:cc(i+1)) = file2mat(obj,int32(1),int32(x(y==i)));
Error in file_array/subsref>subfun (line 98)
t = multifile2mat(sobj,varargin{:});
Error in file_array/subsref (line 65)
t = subfun(sobj,args{:});
Error in meeg/subsref (line 21)
varargout = {double(subsref(this.data.y, subs))};
Error in spm_eeg_invert (line 454)
Y = A{i,m}*D{i}(Ic{i,m},It{i},c(k));
Error in spm_eeg_inv_group (line 208)
D = spm_eeg_invert(D);
Error in spm_eeg_inv_imag_api>group_Callback (line 480)
spm_eeg_inv_group;
Error in spm_eeg_inv_imag_api (line 53)
feval(varargin{:}); % FEVAL switchyard
Error while evaluating uicontrol Callback
Thanks,
Davide
-----Original Message-----
From: Vladimir Litvak [mailto:[log in to unmask]]
Sent: Thursday, 23 February 2012 4:18 PM
To: Rivolta, Davide
Cc: [log in to unmask]
Subject: Re: [SPM] MEG - group source reconstruction
Dear Davide,
If you use individual MRIs you must create the head models individually. If
you have individual fiducials then you must also do the coregistration
individually. If you also choose the forward model and save it you can then
run group inversions repeatedly on these datasets without having to re-do the
head-modelling bit. If you want to script everything you can use the batch
tool for creating head models to create a batch and do 'Save as..' to save it
as an .m file.
You can then modify the batch inputs with your own code and re-run.
This is not M/EEG-specific there must be a lot of explanations about that if
you search in the list archives.
Best,
Vladimir
On Thu, Feb 23, 2012 at 2:48 PM, Rivolta, Davide
<[log in to unmask]> wrote:
> Dear all,
>
>
>
> Is there a documentation (apart from the manual) that explains how to
> perform the "group inversion"?
>
>
>
> For example, do I need to perform the co-registration, Forward model
> and Invert resolution for each subject (and each condition)
> individually and then, somehow, average the results, OR there is a
> quicker way (I am thinking at group inversion) to run all these steps at
once?
>
>
>
> I am using individual MRI.
>
>
>
> Thanks,
>
> Davide
>
>
|