Hello AFNI experts,

I was writing before to ask what metric I could use to summarize the spikiness measure in an ROI after using 3dDespike.

What I did so far is using 3dDespike which gave me an output dataset that includes the spikiness measure 's' for each timepoints of every voxel.
I then used 3dToutcount to calculate the fraction of outliers at each time point in my ROI.

However, I realized just now that this might not be the right function to use as 3dToutcount calculates a trend and MAD across the time series for each voxel and then determines the outliers when a time point is far away from this trend. So in my case it calculates in each voxel a trend over the spikiness measure 's' across all time points and outlier voxels will be timepoints where the spikiness measure deviates from the trend in this voxel.That does not sound like the right thing to do.

What I would like to do instead is to:

1. calculate for each timepoints the fraction of voxels in my ROI where the spikiness measure 's' exceeds a certain threshold. (I think the default threshold in 3dDespike is 2.5)

2. Count for each subject the number of timepoints where this fraction exceeds a certain threshold

Does this sound right and what AFNI function can I use that would allow me to do step 1 and 2?

Thank you very much in advance!
Carolin