afni.nimh.nih.gov
AFNI Message Board
Dear AFNI users-
We are very pleased to announce that the new AFNI Message Board framework is up! Please join us at:
Existing user accounts have been migrated, so returning users can login by requesting a password reset. New users can create accounts, as well, through a standard account creation process. Please note that these setup emails might initially go to spam folders (esp. for NIH users!), so please check those locations in the beginning.
The current Message Board discussion threads have been migrated to the new framework. The current Message Board will remain visible, but read-only, for a little while.
Sincerely, AFNI HQ
We are very pleased to announce that the new AFNI Message Board framework is up! Please join us at:
https://discuss.afni.nimh.nih.gov
Existing user accounts have been migrated, so returning users can login by requesting a password reset. New users can create accounts, as well, through a standard account creation process. Please note that these setup emails might initially go to spam folders (esp. for NIH users!), so please check those locations in the beginning.
The current Message Board discussion threads have been migrated to the new framework. The current Message Board will remain visible, but read-only, for a little while.
Sincerely, AFNI HQ
| October 22, 2014 05:19PM | |
We are very pleased to release the Haskins Pediatric Atlas and Template. This initial release, version 1.0, includes nonlinear and affine templates and corresponding segmentation atlases. Binary distributions following this posting will include these atlas and template datasets. The datasets were derived from T1 anatomical data from an initial population of 75 typically developing subjects, ages 7-12 years old.
The affine template was created from an iterative affine alignment using @auto_tlrc from a rigid equivalent (AC-PC aligned) mean. Using the @toMNI_Qwarpar script, we computed the nonlinear template with the affine template as its starting point and affinely aligned individual subjects and generated new means using a nonlinear fit over progressively smaller neighborhood sizes. Outlier datasets were determined by computing the mean warp deformation from each dataset to all others. Subjects were manually removed based on the mean warp deformation, and the procedure was repeated for both affine and nonlinear templates. The final nonlinear template was affinely aligned to the affine template to correct for general size, and the template origins were repositioned to have the anterior commissure at the (0,0,0) position.
Atlas information was generated for each subject using Freesurfer's automatic segmentation. Segmentation was visually verified for obvious issues. The segmentation was transformed to the template space using the corresponding affine and nonlinear transformations. Probability maps were composed for each region, and a map of the region with the maximum probability was found at each voxel. These maximum probability maps (MPM) were smoothed using a local neighborhood histogram procedure to find the modal region at each voxel.
In AFNI, whereami will show these atlases if the user sets the AFNI_ATLAS_LIST to include the atlas names. A transformation to MNI space was computed using an affine alignment of the nonlinear template to the MNI-152 template. This transformation allows the use of the adult atlases at the same time as the pediatric atlases, and the corresponding MNI coordinates can be computed in the whereami output. Setting the following environment variables in your .afnirc file or in the shell will enable the atlases and spaces to be included in the AFNI whereami functions.
AFNI_ATLAS_LIST "Haskins_Pediatric_Nonlinear_1.0,Haskins_Pediatric_Affine_1.0,CA_MPM_18_MNIA"
AFNI_ATLAS_COLORS Haskins_Pediatric_Nonlinear_1.0
AFNI_TEMPLATE_SPACE_LIST HaskinsPeds,MNI,TLRC
For studies with pediatric subjects, users may want to consider alignment to these templates using AFNI's standard tools such as @auto_tlrc for affine alignment or auto_warp.py, both available within the context of afni_proc.py.
# for affine transformations, either template is appropriate
@auto_tlrc -base HaskinsPeds_NL_template1.0+tlrc -input anat+orig.
or
@auto_tlrc -base HaskinsPeds_aff_template1.0+tlrc -input anat+orig.
# for nonlinear transformations, the nonlinear template will be more useful
auto_warp.py -base HaskinsPeds_NL_template1.0+tlrc -input strip+orig.
While this is still an early phase of the atlas, we wanted to make this available to the general community before publication. More information is available in the following presentations:
http://afni.nimh.nih.gov/pub/dist/edu/latest/afni10_volreg_talairach/
http://afni.nimh.nih.gov/pub/dist/atlases/HaskinsPeds/HaskinsPedsDist_1.0.tgz
In the image below, the left pair of images shows the size of the pediatric template relative to the MNI152_2009a_nonlinear template; the center pair shows the nonlinear atlas and template; the right pair shows the affine atlas and template.
Peter Molfese and Daniel Glen
Edited 4 time(s). Last edit at 10/22/2014 06:31PM by Brian Pittman.
The affine template was created from an iterative affine alignment using @auto_tlrc from a rigid equivalent (AC-PC aligned) mean. Using the @toMNI_Qwarpar script, we computed the nonlinear template with the affine template as its starting point and affinely aligned individual subjects and generated new means using a nonlinear fit over progressively smaller neighborhood sizes. Outlier datasets were determined by computing the mean warp deformation from each dataset to all others. Subjects were manually removed based on the mean warp deformation, and the procedure was repeated for both affine and nonlinear templates. The final nonlinear template was affinely aligned to the affine template to correct for general size, and the template origins were repositioned to have the anterior commissure at the (0,0,0) position.
Atlas information was generated for each subject using Freesurfer's automatic segmentation. Segmentation was visually verified for obvious issues. The segmentation was transformed to the template space using the corresponding affine and nonlinear transformations. Probability maps were composed for each region, and a map of the region with the maximum probability was found at each voxel. These maximum probability maps (MPM) were smoothed using a local neighborhood histogram procedure to find the modal region at each voxel.
In AFNI, whereami will show these atlases if the user sets the AFNI_ATLAS_LIST to include the atlas names. A transformation to MNI space was computed using an affine alignment of the nonlinear template to the MNI-152 template. This transformation allows the use of the adult atlases at the same time as the pediatric atlases, and the corresponding MNI coordinates can be computed in the whereami output. Setting the following environment variables in your .afnirc file or in the shell will enable the atlases and spaces to be included in the AFNI whereami functions.
AFNI_ATLAS_LIST "Haskins_Pediatric_Nonlinear_1.0,Haskins_Pediatric_Affine_1.0,CA_MPM_18_MNIA"
AFNI_ATLAS_COLORS Haskins_Pediatric_Nonlinear_1.0
AFNI_TEMPLATE_SPACE_LIST HaskinsPeds,MNI,TLRC
For studies with pediatric subjects, users may want to consider alignment to these templates using AFNI's standard tools such as @auto_tlrc for affine alignment or auto_warp.py, both available within the context of afni_proc.py.
# for affine transformations, either template is appropriate
@auto_tlrc -base HaskinsPeds_NL_template1.0+tlrc -input anat+orig.
or
@auto_tlrc -base HaskinsPeds_aff_template1.0+tlrc -input anat+orig.
# for nonlinear transformations, the nonlinear template will be more useful
auto_warp.py -base HaskinsPeds_NL_template1.0+tlrc -input strip+orig.
While this is still an early phase of the atlas, we wanted to make this available to the general community before publication. More information is available in the following presentations:
http://afni.nimh.nih.gov/pub/dist/edu/latest/afni10_volreg_talairach/
http://afni.nimh.nih.gov/pub/dist/atlases/HaskinsPeds/HaskinsPedsDist_1.0.tgz
In the image below, the left pair of images shows the size of the pediatric template relative to the MNI152_2009a_nonlinear template; the center pair shows the nonlinear atlas and template; the right pair shows the affine atlas and template.
Peter Molfese and Daniel Glen
Edited 4 time(s). Last edit at 10/22/2014 06:31PM by Brian Pittman.
