xingyu-liu
commented
1 month ago Hi! Thank you for your question. After reviewing the code again, I can confirm that the correct atlases for the analyses are the Acapulco segmentation for each individual:
# {'t1wT2wRatio' : 'cb_acapulco_adult', 'fALFF' : 'cb_acapulco_adult-2mm'}
This means you do not need the Cerebellum-MNIfnirt.txt file.
However, to perform an Acapulco analysis, you will first need to run the Acapulco segmentation on each individual subject. If this step is not feasible for your setup, you might consider using the cb_anat_fsl atlas as an alternative. You could do it by adapting the Cerebellum-MNIfnirt.txt file to match the format of acapulco_adult.txt. Note that the results might be slightly different from using an more accurate individual-level segmentation like Acapulco.
Hi I am sorry to bother you again. Thank you for uploading Cerebellum-MNIfnirt.txt .But now I met a problem, it need 5 columns in cb_tool.py file,for example: atlas_info = pd.read_table(os.path.join(atlas_dir, 'Cerebellum-MNIfnirt.txt'), sep='\t', header=None) atlas_info.columns = ['key','name','lobule','hemi', 'landmark'] atlas_info['landmark'] = atlas_info['landmark'].apply(lambda x: np.array(eval(x))) but this file 'Cerebellum-MNIfnirt.txt ' only has 3 columns : 'roi_mask' 'name' 'hemi'. Is there a new Cerebellum-MNIfnirt.txt file?
Best wishes