Hello everyone,
I am working with a dataset preprocessed using an older version of fMRIPrep (0.4.4), and I am trying to parcellate my signal according to the 68 cortical regions of the Desikan-Killiany atlas. I am looking for the easiest way to do this.
Here are the files I have available:
- Main signal:
sub-XXX_task-rest_bold_space-MNI152NLin2009cAsym_preproc.nii.gz - Brain mask, WM, CSF, and GM probability maps in the same space
- Signal in each participant’s T1w space (I believe):
sub-XXX_task-rest_bold_space-T1w_preproc.nii.gz - Warp file:
sub-XXX_T1w_space-MNI152NLin2009cAsym_warp.h5 aparc+aseg.mgzin the FreeSurfer/mri folder, which I assume is the atlas in the individual T1w space (is that correct?)
I was attempting to use the warp file to transform the atlas into MNI space, but it wasn’t as straightforward as I expected.
My goal is to use the DK-68 atlas for my data while still being able to proceed with my planned denoising steps without having to rerun the entire preprocessing. I am considering either ICA-AROMA+2P or 6M+2P for denoising.
How can I achieve this? Aside from transforming the atlas, I’ve been looking for a DK atlas in MNI space online, but I haven’t been successful. I’ve also considered running the analysis in each participant’s T1w space, but I only have probability masks in MNI space.
Any suggestions would be greatly appreciated.
PS: I noticed there were a few similar questions here before, but I believe they didn’t receive answers.