Dear FSL experts,
Sorry to post it again, but I am not finding a solution for the Melodic's error.
I’m working with a PASL dataset with 90 volumes and I have been running two experiments using Melodic ICA, in order to analyze RSN’s. For the first experiment the input to Melodic was the perfusion signal obtained by surround subtraction (of the pasl time series already pre-processed using Feat-Prestats) and for the second the BOLD signal obtained by surround average (also of the pasl time series already pre-processed).
1) For 4 of my subjects I have run into an error in Melodic that I'm having trouble diagnosing. This happens when the input is the BOLD signal.
/usr/local/fsl/bin/fsl_sub -T 8 -l logs -N feat2_pre -j 4844 /usr/local/fsl/bin/feat /home1/user1/Data/PATIENTS/patient_0008/pairdifference2_series.ica/design.fsf -D /home1/user1/Data/PATIENTS/patient_0008/pairdifference2_series.ica -I 1 -prestats
divide by zero
while executing
"if { [ catch {
for { set argindex 1 } { $argindex < $argc } { incr argindex 1 } {
switch -- [ lindex $argv $argindex ] {
-I {
incr argin..."
(file "/usr/local/fsl/bin/feat" line 119)
</pre><hr>Initialisation<br><pre>
Anyone have an idea of what is causing this?
2) The second question is about the number of resulting IC's. For the BOLD signal I obtain, on average, 25 ICs while for the perfusion signal I have on average only 10 ICs. According to your experience, is it expectable this pronounced difference and is there a reason for this (since the total number of volumes is the same) ?
Thanks in advance for your help!
Regards,
--
Márcia Santos
MSc Student in Biomedical Engineering
Instituto Superior Técnico - Technical University of Lisbon
Portugal
|