On 28 Apr 2014, at 21:47, Norman Benbrahim <[log in to unmask]> wrote:

Hello FSL experts,
I have something of a theoretical question for you. Given a 10 minute resting state scan from one subject, how many RSN's do you expect to find? Let's assume you run a standard EPI, and there is relatively little motion and the data comes out pretty clean.

I would expect probably 50-150 components, about 20-33% of these being RSNs.

What would you do (other than run through standard preprocessing and MELODIC two to three times, using reg_filt along the way) to ensure the best possible results?

That is a good approach, yes - although I think it is rarely needed to run the ICA cleanup more than once. Note that you can use FIX to auto-classify the components for the cleanup - particularly if you train FIX carefully for your protocol.

Cheers.

I know this is fairly vague, but I'm in somewhat of a pickle because I can't tell whether I did something wrong in my analysis, or maybe my data just isn't as clean as I had hoped, so I thought I might be able to pick your brains (pun intended). Thanks!

---------------------------------------------------------------------------
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director, Oxford University FMRIB Centre

FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve

---------------------------------------------------------------------------

Stop the cultural destruction of Tibet