Glad you got it all figured out.  I’m curious though:  was this exercise just 
to understand the various steps internal to FIX (and hcp_fix) or is there some 
more fundamental reason that you can't use those actual scripts for your 
processing?

cheers,
-MH

--
Michael Harms, Ph.D.
-----------------------------------------------------------
Conte Center for the Neuroscience of Mental Disorders
Washington University School of Medicine
Department of Psychiatry, Box 8134
660 South Euclid Ave. Tel: 314-747-6173
St. Louis, MO  63110 Email: [email protected]

From: 
<[email protected]<mailto:[email protected]>>
 on behalf of "Ely, Benjamin" 
<[email protected]<mailto:[email protected]>>
Date: Tuesday, October 4, 2016 at 1:07 AM
To: "Burgess, Gregory" <[email protected]<mailto:[email protected]>>, 
"[email protected]<mailto:[email protected]>" 
<[email protected]<mailto:[email protected]>>
Subject: Re: [HCP-Users] MELODIC denoising vs. released ICA-FIX datasets

Hi Greg,

Thank you very much for your detailed response!

I have now generated a Movement_Regressors.mat file with the correct 24 
parameters (6 rigid body, 6 derivatives, 6 rigid body squared, 6 derivatives 
squared). As I didn't have code handy to run highpass filtering on the movement 
regressors, I based this file on the detrended movement regressors; the HCP's 
highpass filtering is described as "detrending-like", so this should 
(hopefully) give approximately the same result. I also noticed an error in my 
original highpass filtering of the fMRI data that accounts for part of the 
difference I was seeing; the correct sigma is 1389, not 1000 (cutoff = 
2*TR*sigma, so a TR of 0.72s and cutoff of 2000s requires a sigma of 1389; I 
forgot to account for the TR in my original filtering). Between the two, I am 
able to get much closer to matching the HCP's denoised data.

Regarding your second point, I think my setup is removing all of the variance 
in the movement parameters, but not in the noise ICs. fsl_glm, which is how I 
remove the movement parameters, appears to always do full regression. By 
default, though, fsl_regfilt performs partial regression of the specified noise 
ICs. But the difference regarding movement regression would definitely make a 
difference and seems to be one thing that can't be matched using FSL only.

So yes, this was very informative! Thank you again for your help.
-Ely

_______________________________________________
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