I tried to the hierarchical clustering by nilearn these days, but afaik the linkage method is based on distance, how it can finish the clustering step on fMRI data? And can it be suitable for one single parameter data like PET data? And how?
how it can finish the clustering step on fMRI data?
I did not understand your question exactly. May be it depends on number of clusters you defined.
And can it be suitable for one single parameter data like PET data? And how?
Single parameter you mean single subject ?
I said this cause i read the sentence “Ward’s algorithm is a hierarchical clustering algorithm: it recursively merges voxels, then clusters that have similar signal (parameters, measurements or time courses).” from the website of nilearn, so i am curious about what exactly the program will do for “parameters and measurements”?
I must admit that I didn’t tried using parcellations on PET data. I think you can try if PET data can represent like fmri images like 4D.