unc brain atlas:MainPage
From NITRC Wiki
Contents |
Overview
3 human brain atlases have been created at the NeuroImage Research and Analysis Laboratories at UNC-Chapel Hill: a pediatric, an adult and an elderly one. Each atlas has been created from training images by iterative, joint deformable registration into an unbiased average image.
These atlases are a mirror from the ones available to the community on MIDAS.
Usage
These atlases can be used for three purposes:
- Tissue segmentation by atlas-based tissue segmentation algorithms
In order to perform a tissue segmentation of the brain, one can use several atlas-based tools: itkEMS (application part of the NeuroLib, developped at UNC-Chapel Hill), EMSegment (a 3D Slicer module), or other applications. In this case, such algorithms need the following images: grayscale image as well as the tissue segmentation probability maps (white, gray, csf, rest).
- Subcortical structures segmentation
Using the grayscale image, one can perform an atlas-to case registration (affine, B-Spline, fluid registration...). Applying the related transformation to the subcortical structures probability maps will allow the user to perform the segmentation of subcortical structures.
- Lobar analysis
One can apply the same transformation to the parcellation map.
Atlases description
Pediatric atlas
This symmetric atlas has been created using 10 4-year old cases plus mirrored ones. Not only is the T1-weighted image provided (with and without skull), but also tissue segmentation probability maps (white matter, gray matter, CSF, rest), subcortical structures probability maps (amygdala, caudate, hippocampus, pallidus, putamen), and a 25-lobe parcellation map.
This work is funded in part by:
- the National Alliance for Medical Image Computing (NAMIC), a NIH Roadmap for Medical Research, Grant U54 EB005149
Adult atlas
This atlas has been created using 50+ healthy adult cases (between 20 and 59 years old) plus inverted ones. Not only is the T1-weighted image provided (with and without skull), but also tissue segmentation probability maps (white matter, gray matter, CSF, rest) and a 27-lobe parcellation map.
This work is funded in part by:
- the National Alliance for Medical Image Computing (NAMIC), a NIH Roadmap for Medical Research, Grant U54 EB005149
- UNC Neurodevelopmental Disorders Research Center HD 03110.
Elderly atlas
This atlas has been created using 27 healthy elderly cases (60+ years old). Not only is the T1-weighted image provided, but also subcortical structures probability maps (amygdala, caudate, hippocampus, pallidus, putamen), and a 26-lobe parcellation map.
This work is funded in part by:
- the National Alliance for Medical Image Computing (NAMIC), a NIH Roadmap for Medical Research, Grant U54 EB005149
- UNC Neurodevelopmental Disorders Research Center HD 03110.
- UNC's Biomedical Research Imaging Center seed grant (PI Dr. Bonita L. Marks, "Role of Recreational Sport Participation on Cerebral White Matter Integrity in Older Adults") and A&S Interdisciplinary Initiative Grant
- UNC Casilab's grant NIH-NIBIB R01-EB000219
Some of the MR brain images from healthy volunteers used in this paper were collected and made available by the CASILab (led by Dr. Elizabeth Bullitt) at The University of North Carolina at Chapel Hill.
