You should use the multi-run training file if you are classifying multi-run data, as it is a bit better.
Multi-run FIX variance normalizes each run prior to concatenation, and this may not work well with very short runs. I don't think we have tested this yet with runs shorter than 100 timepoints.
For HCP data, we generally required runs to have at least 75% of timepoints to call them "usable" for data release.
Matt.
On 6/9/20, 12:23 PM, "FSL - FMRIB's Software Library on behalf of Dustin Moraczewski" <[log in to unmask] on behalf of [log in to unmask]> wrote:
* External Email - Caution *
Hello -
We are running many of the ABCD subjects through ICA-FIX using fix_multi_run.sh and a compiled matlab script:
fix_multi_run.sh [log in to unmask]@[log in to unmask] 2000 fix_proc/task-rest_concat TRUE HCP_hp2000.RData
However, for some subjects we are getting the following error:
4.1.3/ICAFIX/scripts/Compiled_functionhighpassandvariancenormalize/run_functionhighpassandvariancenormalize.sh MCR/v93 0.800000 2000 '\''task-rest02'\'' '\''/usr/local/apps/connectome-workbench/1.4.2/wb_command'\'' '
Subscript indices must either be real positive integers or logicals.
Error in icaDim>FitWishart (line 257)
Error in icaDim (line 100)
Error in functionhighpassandvariancenormalize (line 165)
MATLAB:badsubscript
One thing we noticed is that runs that failed seem to have very few timepoints. For example, a run that has 46 timepoints fails where a full run (ie 380 timepoints) succeeds. This is fine, I don't really want to include a 46-timepoint run anyway. However, my questions are:
1. Is the high pass failing because of too few timepoints? If not, what else could cause this function to fail at this point?
2. If it is too few timepoints, how would you recommend we assess whether or not a run will fail? Given our high pass and how fslmaths -bptf works, is there a number of volumes that we can say definitively that this function needs to succeed?
Our TR length is 0.8 seconds, if that's relevant.
Thank you for your guidance!
Best,
Dustin
########################################################################
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
This message was issued to members of www.jiscmail.ac.uk/FSL, a mailing list hosted by www.jiscmail.ac.uk, terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/
________________________________
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.
########################################################################
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
This message was issued to members of www.jiscmail.ac.uk/FSL, a mailing list hosted by www.jiscmail.ac.uk, terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/
|