 |
 |
Dear all, I would like to clarify whether my FEAT analysis is correct or not. I would appreciate it very much if you could give me comments and suggestions. We aim to identify a linear relationship between a clinical score and the degree of functional connectivity (FC). First, we conducted seed based FC analysis at first level in patients. Second, in order to identify a linear relationship between a clinical score and the degrees of FC, I conducted higher level analysis as instructed in “Single-group Average with Additional Covariate” (http://fsl.fmrib.ox.ac.uk/fsl/fslwiki/GLM#Single-Group_Average_with_Additional_Covariate). I set EV1 as group mean, and EV2 as a clinical score, and performed voxel corrected analysis (p<0.05) with fixed effect model. As a result, we obtained significant area (A) with a positive linear relationship with the clinical score. Then, I tried to plot the relationship between the clinical score and the degree of FC. Taking the statistical peak voxel of area (A) as a mask, I extracted COPE values in individual patients. I plotted, and what I have found was that the clinical score did not seem to correlate with the COPE value. Are these procedures correct to identify a linear relationship between a clinical score and the degree of FC? I would appreciate if anyone could give me helpful comments on this issue. Best wishes, Atsushi Shima Graduate student Department of Neurology, Kyoto University Graduate School of Medicine Japan