Good morning/ afternoon,
We have a sample subject with 1000 time points pre-processed in DPARSF (information included) that we are trying to conduct MELODIC ICA on. However, we are encountering a problem during the Prestats phase/ mask retrieval.
We understand the theoretical and technical underpinnings and it makes sense as to what FSL MELODIC is trying to do here and running into problems, just trying to find out exactly what settings to use (or not to use) within MELODIC so we can avoid this error and complete the analysis.
Attachments:
1. Sample image input to MELODIC, part of 4D and pre-processed in DPARSF
2. report_log from MELODIC
Would highly appreciate all feedback/ input. This is causing a road block to entire analysis. We have tried several settings in MELODIC with the same error (given below and in the log file attached).
Many thanks!
Syed
-----------------------------------------------------------------------------------------------------------------------
Supplementary Information:
Raw Data:
1000 Time Points (1000 dcms, No anatomical images specified)
Error:
/export/apps/fsl/bin/fslstats prefiltered_func_data_bet -p 2 -p 98
-4.667094 4.786029
/export/apps/fsl/bin/fslmaths prefiltered_func_data_bet -thr -3.7217817 -Tmin -bin mask -odt char
/export/apps/fsl/bin/fslstats prefiltered_func_data_mcf -k mask -p 50
0.000000
/export/apps/fsl/bin/fslmaths mask -dilF mask
/export/apps/fsl/bin/fslmaths prefiltered_func_data_mcf -mas mask prefiltered_func_data_thresh
/export/apps/fsl/bin/fslmaths prefiltered_func_data_thresh -Tmean mean_func
/export/apps/fsl/bin/susan prefiltered_func_data_thresh 3.5003205 2.12314225053 3 1 1 mean_func 3.5003205 prefiltered_func_data_smooth
/export/apps/fsl/bin/fslmaths prefiltered_func_data_smooth -mas mask prefiltered_func_data_smooth
divide by zero
while executing
"if { [ catch {
for { set argindex 1 } { $argindex < $argc } { incr argindex 1 } {
switch -- [ lindex $argv $argindex ] {
-I {
incr argin..."
(file "/export/apps/fsl/bin/feat" line 119)
DPARSF Pre-Processing:
-Time Points = 1000
-TR (sec) = 0.6
-Realign
-Normalize
-Bounding Box remains same [-90 -126 -72; 90 90 108]
-Voxel Size: [2 2 2]
-EPI templates
-Data without smooth
-Detrend
-Filter (0.01-0.1)
-Default mask
-Regress out nuisance covariates:
6 head motion parameters
White matter signal
Cerebrospinal fluid signal
Global mean signal (if global mean sig regression is preferred only)
|