Dear all,
I hope this email finds you well..
I have a question about what kind of processing should be done before matching the fMRI data and the atlas. I got a suggestion that “additional planes of zeros should be firstly stripped off before changing the atlas to match the data”. If the Harvard-oxford atlas and the preprocessed functional image have different number of additional planes of zeros. What kind of processing should be done before matching them.
The detailed information is shown as follows:
I downloaded the preprocessed functional image that is processed using C-PAC pipeline, with frequency filtering and global signal regression, from the ABIDE website. (I used the following command to download the preprocessed functional image:
python C:/download_abide_preproc.py -d func_preproc -p cpac -s filt_global -o D:/ABIDE1_File
I tried to find out the Harvard-oxford atlas having overlapping volumes with the preprocessed functional image that is processed using C-PAC pipeline. However, I only find the Havard-oxford atlas for NIAK pipeline with non- overlapping volumes with the preprocessed functional image from the following link of ABIDE http://preprocessed-connectomes-project.org/abide/Pipelines.html
I loaded the Harvard-oxford atlas into Matlab by using “load_nii” function, and found that more than 16 planes are the planes of zeros in each of x, y, z direction. I also loaded the preprocessed functional image (e.g., Yale_0050626_func_preproc.nii). There are also a few planes of zeros for the preprocessed functional image. But the number of the planes of zeros in the preprocessed functional image is different from the Harvard-oxford atlas.
Best Regards
Zhongxing Zhou