CSO mask in MNI space does not convert well into individual space

[GinaGriffioen]

I have received an AAL based mask for CSO from Yale, but when I try to convert it into individual space, it it totally off.

The mask can be found here: \130.229.41.50\alfred\Atlas_Images\vff\Gina2\CSO mask Yale\C_semiovale_2ml

I copied the matlab file, which was created when I normalised WM to create my own CSO mask from WM and renamed it to acronym_CSO_Yale

In my study lay-out I use: MR { normaliseImage8 $normalized_mr_ana_dir/average_nsr_fs_seg_WM_er5_thr09bin iy_r|MR_crop INTERP Nearest neighbour normaliseImage8 $normalized_mr_ana_dir/CSO_Yale iy_r|MR_crop INTERP Nearest neighbour }

the command works for the first line (which is my own population based CSO mask in MNI space), but not for the second.

The CSO is off, both in SPM and in Vinci. The TACs are capturing almost no radioactivity (which makes sence, since the ROI is outside of the brain)

compare Nifti header from mask in MNI space to the mask in individual space:

The qoffset_y is positive in MNI space and negative in individual space. Maybe that is why it is so different?

What I tried already to solve the issue:

• used a different matlab file (from MR_crop)
• flipped the orientation of the CSO mask

Any other idea's?