Number of components depends on a large number of things including length of 
run, TR, voxel resolution, brain state, and amount of motion.  In general as 
scans improve/have more information in them you will get more total components. 
 Also as you have more motion in a scan, this will increase the number of bad 
components.  As successful ICA run separates good and bad components cleanly so 
that the bad components can be regressed out without removing neural signal.  
In general this is easier to do in better quality data and there is no ratio of 
good to bad that is expected.

Matt.

From: 
<[email protected]<mailto:[email protected]>>
 on behalf of "Theis, Nicholas" <[email protected]<mailto:[email protected]>>
Date: Wednesday, August 1, 2018 at 12:05 PM
To: "[email protected]<mailto:[email protected]>" 
<[email protected]<mailto:[email protected]>>
Subject: [HCP-Users] Number of "good" vs "bad" independent components 
classified by FIX


Dear HCP community,


While the number of "good" and "bad" components certainly depends on the fMRI 
paradigm, such as a specific task versus a resting state scan, as well as the 
quality of the scan and similarity to training data (I'm using 
HCP_hp2000.RData), is there in general an expected value for the number of 
"good" components in a successful ICA?


For instance, on a rsfMRI scan should MELODIC produce, say, 100 components, 10 
of which are classified as "good" by FIX, 80 as "bad" and 10 as "unknown"?  Or 
is the breakdown usually closer to 50% "good", 50% "bad" - in your experience?  
Or is there really no heuristic as far as a proportion of good/bad/unknown 
components in a successful ICA of fMRI data?


Thanks for your time,

Nick Theis

_______________________________________________
HCP-Users mailing list
[email protected]<mailto:[email protected]>
http://lists.humanconnectome.org/mailman/listinfo/hcp-users

________________________________
The materials in this message are private and may contain Protected Healthcare 
Information or other information of a sensitive nature. If you are not the 
intended recipient, be advised that any unauthorized use, disclosure, copying 
or the taking of any action in reliance on the contents of this information is 
strictly prohibited. If you have received this email in error, please 
immediately notify the sender via telephone or return mail.

_______________________________________________
HCP-Users mailing list
[email protected]
http://lists.humanconnectome.org/mailman/listinfo/hcp-users

Reply via email to