AFNI-filter problems

1

Hi all,
I’m a freshman in MRI data preprocessing. I perform filter (0.01-0.1) and detrend as follows, but my results seems problematic as attached (both in human and monkey), can anyone tell me why and how to fix this?

3dTproject -prefix filt.nii.gz -passband 0.01 0.1 -input rfMRI.nii.gz
3dTstat -mean -prefix filt_mean.nii.gz filt.nii.gz
3dDetrend -polort 2 -prefix dt.nii.gz filt.nii.gz
3dcalc -a filt_mean.nii.gz -b $dt.nii.gz -expr ‘a+b’ -prefix Final.nii.gz

The first row are original figures and the second are that filtered

screenshot_1
screenshot_11339×1098 155 KB

Thank you so much

Jewel

2

It’s not clear to me what you are trying to do here. 3dTproject includes detrending and demeaning in its output. By default, a polort of 2 is used (a second order polynomial) baseline fit. The output is a 3D+time dataset with a mean across time that is near zero at every voxel.