About hierarchical parcellation: how to parcel the fMRI data based on anatomical structures?

Hi everyone,
I had used the nilearn package to do hierarchical clustering on fMRI data, but after I check the results and found that some of the clusters had much more voxels than others. And I tried to do this based on anatomical structures, just like the same thing in this article:
“Liu X, Ward D, Binder JR, Li S-J, Hudetz AG (2014) Scale-Free Functional Connectivity of the Brain Is Maintained in Anesthetized Healthy Participants but Not in Patients with Unresponsive Wakefulness Syndrome. PLoS ONE 9(3): e92182. doi:10.1371/journal.pone.0092182”
As descried in “Network partitioning and evaluation of scale-free statistics”, “Second, hierarchical clustering was conducted separately each time with the preprocessed BOLD time series of all voxels included in each of the 116 anatomical structures.”
Could anyone tell me if it is possible to finish it in nilearn?
Thanks for your attention!