help
help > RE: voxel-to-voxel level connectivity matrix
Mar 14, 2016 07:03 PM | Alfonso Nieto-Castanon - Boston University
RE: voxel-to-voxel level connectivity matrix
Dear Pravesh,
Yes, that is a good idea. I imagine the differences in ROI numbers arise from the newer SPM12 TPM file defining a somewhat more conservative gray matter mask compared to the original SPM8 grey.nii mask, rather than differences in their original voxel sizes, but you are right that it would be most appropriate to use a parcellation derived from the same gray-matter mask that has been used in the normalization step (so that the resulting parcellation best captures/covers the most likely extent of gray matter voxels after normalization). And regarding your question about bounding box definitions, I do not believe it would be necessary to modify this (or anything related to different voxel sizes in the newer masks) since the buildP.m function should be relatively insensitive to these choices (as longs as the entered bounding box sufficiently covers the expected gray matter voxel locations)
Hope this helps
Alfonso
Originally posted by Pravesh Parekh:
Yes, that is a good idea. I imagine the differences in ROI numbers arise from the newer SPM12 TPM file defining a somewhat more conservative gray matter mask compared to the original SPM8 grey.nii mask, rather than differences in their original voxel sizes, but you are right that it would be most appropriate to use a parcellation derived from the same gray-matter mask that has been used in the normalization step (so that the resulting parcellation best captures/covers the most likely extent of gray matter voxels after normalization). And regarding your question about bounding box definitions, I do not believe it would be necessary to modify this (or anything related to different voxel sizes in the newer masks) since the buildP.m function should be relatively insensitive to these choices (as longs as the entered bounding box sufficiently covers the expected gray matter voxel locations)
Hope this helps
Alfonso
Originally posted by Pravesh Parekh:
Dear Dr. Alfonso,
Thank you for including this wonderful utility inside Conn. Your contributions and advise are a lot of help (especially to beginners like me)! Thank you :)
Regarding the use of buildP.m file, I was wondering if its a good idea to replace the 'grey.nii' with the newer TPM.nii (first volume)? Doing the same results in a slightly different number of parcels: 270, 360, 504, 738, 1154, 1844 for 20, 18, 16, 14, 12, and 10 mm SideLengths respectively. This is most likely because the two images are different in dimensions and slightly different in voxel sizes too (grey.nii being 2mm while TPM,1 being 1.5mm isotropic).
I am asking this because the newer SPM (12) routines use the TPM.nii for steps like normalization and segmentation (and so does Conn), should the ROIs be made from the same maps? Or would it not make much of a difference either way? If yes, how would the bounding box definition change?
Thanks and Regards
Pravesh
Thank you for including this wonderful utility inside Conn. Your contributions and advise are a lot of help (especially to beginners like me)! Thank you :)
Regarding the use of buildP.m file, I was wondering if its a good idea to replace the 'grey.nii' with the newer TPM.nii (first volume)? Doing the same results in a slightly different number of parcels: 270, 360, 504, 738, 1154, 1844 for 20, 18, 16, 14, 12, and 10 mm SideLengths respectively. This is most likely because the two images are different in dimensions and slightly different in voxel sizes too (grey.nii being 2mm while TPM,1 being 1.5mm isotropic).
I am asking this because the newer SPM (12) routines use the TPM.nii for steps like normalization and segmentation (and so does Conn), should the ROIs be made from the same maps? Or would it not make much of a difference either way? If yes, how would the bounding box definition change?
Thanks and Regards
Pravesh
Threaded View
| Title | Author | Date |
|---|---|---|
| Yifei Zhang | Jul 24, 2014 | |
| Alfonso Nieto-Castanon | Jul 29, 2014 | |
| Alfonso Nieto-Castanon | Jul 29, 2014 | |
| Pravesh Parekh | Feb 29, 2016 | |
| Alfonso Nieto-Castanon | Mar 14, 2016 | |
| Pravesh Parekh | Mar 14, 2016 | |
| Yifei Zhang | Jul 29, 2014 | |