Hi Jeffery,

We developed this for high motion data and it performed well in an overt speech data compared to a number of other methods:

https://www.ncbi.nlm.nih.gov/pubmed/26416652 http://www.homepages.ucl.ac.uk/~ucjttie/index.html

There is also scan nulling / artrepair / afni despike / ICA based fsl tools.

David

From: SPM (Statistical Parametric Mapping) [mailto:[log in to unmask]] On Behalf Of Jeffrey West
Sent: 02 February 2017 19:45
To: [log in to unmask]
Subject: [SPM] question regarding motion correction for severe movement

Hello.

I have a current study in which the subjects presented with a significant amount of movement during the fMRI task scan.  The scan consisted of 4 runs for a total of 917 volumes collected.  While the movement was not constant across all 917 volumes, there were specific volumes with motion from 5mm - 15mm in one direction, as per realignment preprocessing output.  I am regressing out the motion parameters in the first level models for each subject, but I am still concerned this is not enough to correct for the high level of movement, and I do not want to discard the subjects with motion over 4mm's.

My question is:  Are there other options beyond realignment and regressing out the motion parameter, and excluding subject for high motion, are there other options for dealing with high levels of motion during fMRI task?

thank you in advance for any information regarding this matter.

Jef

Jeffrey West, M.A.
Research Analyst
Maryland Psychiatric Research Center
Baltimore, Maryland 21228-0247
Phone: 410-402-6018
email: [log in to unmask]