You could try using compcorr for WM and CSF signals. It takes the first 3 (or whatever you choose) PCAs of those signals are regressors, and may be a bit better at removing motion artifacts.
You can also maybe censor or reinterpolate bad segments of the data with a lot of motion noise, though if I understand your analysis correctly, this could be problematic for you.
But nothing you can do will completely remove motion. If possible, you may need to cull your data set, keeping only those with the least catastrophic signals. I recently worked a data set with a lot of task correlated motion, We ran MELODIC with reg_filt to remove noise components, and even with censoring a few bad TRs I had to reject 8 or 28 people entirely.
Bad data is worse than no data at all.
Good luck.
Colin Hawco, PhD
Neuranalysis Consulting
Neuroimaging analysis and consultation
www.neuranalysis.com
[log in to unmask]
-----Original Message-----
From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On Behalf Of Marie-Eve Hoeppli
Sent: April-28-16 1:53 PM
To: [log in to unmask]
Subject: [FSL] dealing with correlations between movement parameters and regressors of interest
Hi,
I am running some analyses on BOLD fMRI multiband sequences. We recorded continuous ratings during these sequences. I first downsampled these ratings at a rate of 250ms. I then demeaned them and used those ratings as regressors of interest in my GLM model.
I ran a two-step analysis:
1) feat with only MCFLIRT and registration turned on, followed by extraction of CSF-WM regressors using the feat outputs
2) feat with Fieldmap correction, spatial smoothing, no HPF and CSF-WM regressors as nuisance regressors I have some interesting signal, but also a lot of noise, mostly posterior. The noise really looks like something due to movements. So I ran some correlations between the motion parameters I got out of step 1 described above and my regressors of interest (downsampled at the same rate as the motion parameters this time). I obtained some low to moderate correlations (average r =.4) . I also checked for correlations between the regressors of interest and a linear drift and I obtained moderate to high correlations (average r = .6).
I had previously tried to denoise the data using Melodic and HPF, but then there was close to no signal left.
I would be extremely grateful if someone could advise me on how to denoise my data without losing my signal.
Thank you.
Best,
Marie-Eve
|