Thanks, Eugene,
we tried both single-subject and group ICA. Data consist of 150 time points. The basic question is how to assess the well-known resting-state networks (about 8, see for example Cole, Smith & Beckmann 2010, Front Sys Neurosci) for each individual subject (in native space) separately. Now we let Melodic estimate the number of IC. Perhaps one way would be to pick the interesting networks visually from the pool of 40-50 IC?
Thanks for further suggestions,
Volker
|