I want to automatize my fMRI time-series extraction for DCM analysis as I have many subjects and 3 tasks and it is very time consuming to use the GUI to extract them one by one!

I extract the time series according to these steps:

1. I define an F-contrast for the regions of interest.

1. Select the spm file and choose the T-contrast I am interested in

2. Apply no mask and select p<0.05 uncorrected, k=0

3. The i select a small volume to use a mask file '.nii' for the specific region and then select the coordinates with highest T test MNI

4. Select the eigenvariate and adjust the F-contrast for regions of interest.

5. Use a sphere of 8 mm for time series extraction

My main problem is defining where I select the small volume and load and .nii file to jump to the coordinates with highest t-value and these values changes for every VOIs. I would be so grateful if anyone help me with this.

To help, I have added a page to the SPM Wiki describing the procedure for extracting timeseries using the SPM batch. Once you’ve got it working using the batch GUI for a single subject, you can then script it to run over subjects. Please let me know if anything is unclear. https://en.wikibooks.org/wiki/SPM/Timeseries_extraction .

Best

Peter