Dear Freesurfer experts

I want to extract "activation timeseries (BOLD signal changes in each time point) " from a predefined ROI  in a single subject task fMRI data. Previously, I extracted these values from fmcpr.sm0.self.lh.nii.gz  by using mri_segstats, but the data seems to be too noisy. 

I searched the forum and found this solution (mri_glmfit --y fmc.nii --qa --glmdir qa.glm --save-yhat This will detrend the time series (removing mean, linear, and quadratic 

trends). With --save-yhat, one of the outputs will be yhat.mgh (the detrended time series). If  you have a task, the task will still be in there.https://www.mail-archive.com/freesurfer@nmr.mgh.harvard.edu/msg18901.html

I ran this command: mri_glmfit --y fmcpr.nii.gz --qa --glmdir qa.glm --save-yhat

and then this command to extract the timeseries:  mri_segstats --i yhat.mgh --slabel arap_7T_all lh $SUBJECTS_DIR/arap_7T_all/lh.label --id 1 --avgwf-remove-mean --avgwf ./lh.01.txt

I got this error: ERROR: dimension mismatch between input volume and seg  input 174 174 51  seg   148565 1 1

I would like to know if this is the right way to extract timeseries from a task data. and if yes, how can I fix the error.

Thanks a lot

Mona