Applicable to connectivity analyses?

[lrq3000]

I've read the paper and from what I understand neuropower was designed for activity analyses using random field theory such as those done in SPM. But can neuropower work with other kinds of fmri analyses such as seed-based connectivity analysis, or ICA? Let's say instead of using an activation map, I use a connectivity map (and the seed map) in neuropower, then would the result be correct (ie, a correct power analysis of a seed-based connectivity analysis)?