Dear all,
We tried to run Second level Feat analysis (FSL 6.0.1) using Flame1 + outlier de-weighting option and the following experimental design: Single-Group Average with an Additional (demeaned) Covariate, cluster-wise thresholding Z=2.3, P=0.05. Contrasts: 1 0; -1 0
Unfortunately FSL Feat fails with the following error messages:
- "Excessive number of outliers detected. Something has gone wrong."
- "Attempted to multiply images of different sizes", after the command: /usr/local/fsl/bin/fslmaths stats/zstat1 -mas mask thresh_zstat1
The second error seems to occur due to some missing axial slices in zstat1, which is probably the consequence of failed outlier detection in those slices (i.e. first error message).
When we inspect zstat1 by fsleyes, missing slices are quite obvious. The same problem was already posted by others, but no answer until now: https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=FSL;81a5b119.1908
Without outlier de-weighting the analysis runs successfully. FSL FEEDS runs successfully on the computer, so it seems that the above error isn't related to the main functions of FSL. Our paradigm is quite robust (it's a block-design language paradigm) and the error is present with another (amygdala-specific) paradigm as well, so it seems that the error is not specific to the certain paradigm.
Does anybody have any idea how to solve this issue?
Best,
Gabor
########################################################################
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
|