Dear Sadjad
To make your life easier, I suggest you create an SPM with the batch. You don't need to provide it with any data - its only function is to convert the timing from seconds to microtime, as expected by SPM. Open the batch editor, then click SPM -> Stats -> fMRI model specification (design only). You can automate the running of this batch in your scripts if needed.
It will be fine if the confound term is just a [n x 1] column vector of ones, where n is the number of scans.
Let me know if you need further help.
Best
Peter
-----Original Message-----
From: Sadjad Sadeghi [mailto:[log in to unmask]]
Sent: 14 December 2016 17:36
To: [log in to unmask]; Zeidman, Peter <[log in to unmask]>
Cc: Sadjad Sadeghi <[log in to unmask]>
Subject: Re: DCM on simulated data
Dear Peter and SPM experts,
I have generated some synthetic fMRI data using the DCM equations (solving ODEs of neuronal and hemodynamic equations and then adding noise to BOLD signal) and now I want to generate the DCM based on my simulated data. Could you please give more information about constructing the SPM.mat file? I am more confused about how I can define the confound term (DCM.Y.X0). what should I use as the confound term in my generated data and how can I find the size I should use for it? Thank you so much in advance.
Best Regards,
Sadjad
|