Hi folks,
I'm new to this kind of analysis, so I would like to be sure that it 's all ok.
I have two groups of subjects (CTR - n=20 + PAT n=22).
I have followed the TBSS pipeline with the options tbss_2_reg -T, and tbss_3_postreg -S, as suggested in the guide.
Following, I have used the command "design_ttest2 design 20 22" and randomise to perform my t-test groups comparison.
I have attached the images after the tbss_fill command.
When I use the function "cluster" from the Terminal, I get a report with only one finding for the contrast one (that should be CTR FA values > PAT, right?).
Anyway, the report is:
Cluster Index Voxels 1-p-MAX 1-p-MAX X (vox) 1-p-MAX Y (vox) 1-p-MAX Z (vox) 1-p-COG X (vox) 1-p-COG Y (vox) 1-p-COG Z (vox)
1 78417 0.998 84 82 20 89.6 106 85.1
Now, when I see to the region X 84, Y 82, Z 20 on my tbss_fill, the corresponding finding on the Harvard-Oxford subcortical Atlas is a region 97 % Brain Stem (see figure).
It is correct or regions so down on my acquisition could be also a bias? (this is an expected finding for my setting).
When I use the command:
fslview mean_FA -b 0,0.6 mean_FA_skeleton -l Green -b 0.2,0.7 tbss_fill -l Red-Yellow
other regions appear in red-yellow on the skeleton (cingulum, precuneus and others). This means that other "fascicles" have altered FA values compared to my control dataset but these alterations are not significant?
Sorry for my questions, I hope you could help me,
Best regards,
Carlo
|