Atlases

Atlases

Atlas Overview

An atlas of the brain allows us to define its spatial characteristics. Where is a given structure; relative to what other features; what are its shape and characteristics and how do we refer to it? Where is this region of functional activation? How different is this brain compared with a normal database? An atlas allows us to answer these and related questions quantitatively. Brain atlases are built from one or more representations of brain. They describe one or more aspects of brain structure and/or function and their relationships after applying appropriate registration and warping strategies, indexing schemes and nomenclature systems. Atlases made from multiple modalities and individuals provide the capability to describe image data with statistical and visual power.

Brain atlases are built from one or more representations of brain. They describe one or more aspects of brain structure and/or function and their relationships after applying appropriate registration and warping strategies, indexing schemes and nomenclature systems. Atlases made from multiple modalities and individuals provide the capability to describe image data with statistical and visual power.

An atlas can take on many forms, from descriptions of structure or function of the whole brain to maps of groups or populations. Individual systems of the brain can be mapped as can changes over time, as in development or degeneration. An atlas enables comparison across individuals, modalities or states. Differences between species can be catalogued. But in most cases, the value added by brain atlases is the unique and critical ability to integrate information from multiple sources. The utility of an atlas is dependent upon appropriate coordinate systems, registration and deformation methods along with useful visualization strategies. Accurate and representative atlases of brain hold the most promise for helping to create a comprehensive understanding of brain in health and disease.

Disclaimer: All Brain atlases available from this site are distributed in good faith and to the best of our knowledge accurately represent brain structure, function and physiology. However, everyone downloading and using any of these atlases agrees to use them under the usage terms and acknowledges that the use of these atlases is at their own risk. Always check the accuracy of any results you obtain using these atlases. The atlases are NOT FDA Approved for clinical/health applications.

Available Atlases

Alzheimer's Disease Template

The AD atlas is designed to reflect the unique anatomy and physiology of the clinical subpopulation suffering from Alzheimer's disease. The atlas acts as a quantitative framework that correlates the structural, metabolic, molecular and histologic hallmarks of the disease. Additional algorithms are described that use information stored in the atlas to recognize anomalies and label structures in new patients. With the information on group anatomical variability, disease-specific atlases are a type of probabilistic atlas specialized to represent a particular clinical group. The resulting atlas can identify patterns of altered structure or function, and can guide algorithms for knowledge-based image analysis, automated image labeling, tissue classification, and functional image analysis. Since current brain templates poorly represent the anatomy of this clinical population, the resulting atlas offers a framework to investigate the disease. Pathological change can be tracked over time, and disease-specific features resolved. Rather than simply fusing information from multiple subjects and sources, new mathematical strategies are introduced to resolve group-specific features not apparent in individual scans. Corrected for 2015.

CCB Test Data Archive

This Archive is composed of three main groups: children, adults, and elderly subjects, which are further subdivided into Normal Controls and Diseased populations. The disease varies per group, with diseased children, suffering from ADHD, the adults, suffering from Alzheimer's, and the elderly, having dementia. Under each of these directories, both male and female subjects can be found, each containing at least 3 subjects. To be inclusive of variations most commonly found in data-sets and potential effects which they may have on existing programs, work was done in both Native (scanner) space and after alignment to ICBM space (12 parameters). A diagrammatic description of the organization of the data in this archivecan be found at .... We started with the raw volumes of 36 subjects and ended with close to 1,400 volumes of processed data.

Future efforts will involve the delineation of other sub-cortical structures, the acquisition of one subject having gone through multiple protocols and sequences, tissue classification using visual verification and manual touching up of each slice by an individual, in an effort to ensure that tissue is being classified correctly.

Future efforts will involve the delineation of other sub-cortical structures, the acquisition of one subject having gone through multiple protocols and sequences, tissue classification using visual verification and manual touching up of each slice by an individual, in an effort to ensure that tissue is being classified correctly.

Chinese_56 Atlas

Chinese brain Atlas is an average brain template composed of high quality brain MRI data from 56 Chinese young subjects. Seven additional Chinese brains were registered to both ICBM152 and the Chinese_56 atlas. It is found that there is more deformation required to register the additional Chinese brains to the ICBM152 than to the Chinese_56. Thus the Chinese brain template (Chinese_56) better represents the shape and size of the Chinese population.

Cytoarchitectonic Maps

For the localization of activated regions identified by functional imaging an overlay with an anatomical atlas is required. However, there is considerable intersubject structural variability, caused by different macroscopical anatomy, as reflected by a different sulcus pattern, and by different location and extent of cortical areas, which can only be defined microscopically. Therefore, only the spatial normalization of the macroscopical anatomy of an MRI brain dataset to a common reference system permits statements about the structural intersubject variability of architectonic areas and their correlation to functional activations.

In Juelich and Duesseldorf (http://www.fz-juelich.de/ime/) cytoarchitectonic areas have already been analyzed and spatially normalized to ECHBD space. The ICBM 452 space is used within the ICBM consortium and large parts of the neuroscientific community for functional data analysis. Maps of the motor and somatosensory cortex, the auditory cortex, the visual cortex and Broca’s region, and fiber tracts have already been published. However, within SPM, widely used for functional data analysis, brains are normalized to ICBM 452 space. Preceding studies showed the benefit of an overlay of cytoarchitectonic maps and SPM data. Therefore the transformation of the cytoarchitectonic maps into the MNI space is desirable for an easy structural-functional analysis.

View the ICBM Cytoarchitectonic Mappings Table

Fetal Brain Atlas

Fetal brain atlas were built from 7.0T MRI data of 34 postmortem human fetal brains with gestational ages ranging from 15 to 22 weeks. This atlas dataset include atlases of each gestational week and the overall atlas covering the whole period. Advanced normalization tools (ANTS) was used to built these symmetric group-wise atlases. Tensor-based morphology (TBM) and local shape analysis (LSA) were performed based on the atlases. The dataset includes:

• atlases of each week and the overall atlas
• extracted surface of each atlases
• Jacobian determinant of transformation during atlas built
• local shape analysis results after surface registration

Human Atlas

The Human atlas is a spatially accurate, high-resolution three-dimensional (3D) volume of brain anatomy from cryosectioned whole human head. The head of a 76 year old female cadaver was cryosectioned on a heavy duty cryomacrotome (PMV, Stockholm Sweden) modified for quantitative digital image capture. Data were placed into the Talairach coordinate system to create a volume of brain anatomy for atlas reference. The spatial resolution of the original digitized images provided sufficient anatomic detail to clearly delineate gray and white matter and neural structures, including major fiber pathways, subthalamic nuclei, and laminae. Animation proved useful in the conveyance of system anatomy as structures are shown traversing through the neuroaxis. Postmortem cryosectioning paired with this computerized presentation allowed the complete 3D volume data to be distributed and shared as an educational, clinical, and research resource.

ICBM 452 T1 Atlas

The ICBM452 atlas is an average of T1-weighted MRIs of normal young adult brains. The space the atlas is in is not based on any single subject. Instead it is an average space constructed from the average position, orientation, scale, and shear from all the individual subjects. The atlas, therefore, is both an average of intensities and of spatial positioning.

ICBM DTI-81 Atlas

The ICBM DTI-81 Atlas is a stereotaxic probabilistic white matter atlas that fuses DTI-based white matter information with an anatomical template (ICBM-152). This atlas is based on probabilistic tensor maps obtained from 81 normal subjects acquired under an initiative of the International Consortium of Brain Mapping (ICBM). The subjects were normal right-handed adults ranging from 18 to 59 years of age. A hand-segmented white matter parcellation map was created from this averaged map. This map can be used for automated white matter parcellation. The precision of the affine-based image normalization and automated parcellation was measured for a group of normal subjects using manually defined anatomical landmarks.

ICBM Probabilistic Atlases

The ICBM452 atlas is an average of T1-weighted MRIs of normal young adult brains. The space the atlas is in is not based on any single subject. Instead it is an average space constructed from the average position, orientation, scale, and shear from all the individual subjects. The atlas, therefore, is both an average of intensities and of spatial positioning.

ICBM T2 Atlas

The ICBM T2 50 atlas is an average of the T2-weighted MRIs of 50 normal young adult brains (equal gender distribution). The space the atlas is based on is the ICBM 452 deterministic atlas and not on any single subject. It represents the mean volume constructed from the average position, orientation, scale, and shear from all 50 individual subjects. The atlas, therefore, is both an average of T2 intensities as well as spatial positioning.

ICBM Template

The purpose of the ICBM brain template is to provide a reference that includes both a set of coordinates and the associated anatomical labels. The localization of functional imaging results on averages of large number of brain MRI volumes does not provide detailed anatomical information. Alternatively, the spatial registration and normalization of brain images from different individuals to a single high-resolution labeled template recognizes the topographically relevant anatomical substrate for a specific activation.

The ICBM (International Consortium for Brain Mapping) high-resolution single subject template is aligned with the individual subject T1 whole brain MRI provided. An AIR nonlinear warp, a 5th order polynomial, is used for this fit. The demarcated labels on the template are then resampled through the warp transform to the subject MRI. A nearest-neighbor resampling is used to ensure the labels stay as labels. The 0.125mm3 resolution of the template helps ensure that the proper label, at least according to the transform, is used in the resampling.

The volumes are in minc format and can presently be visualized using the program Display. To view both the structural MRI and the labels launch the program typing Display icbm_template.mnc -label icbm_labels_corrected.mnc. The opacity of the labels can be set in the Colour Coding menu. The number of each label appears at the bottom left of the orthogonal views window. This number corresponds to a unique structure as listed in the table provided (labels_text).

LPBA40

The LONI Probabilistic Brain Atlas (LPBA40) is a series of maps of brain anatomic regions. These maps were produced from a set of whole-head MRI of 40 human volunteers. Each MRI was manually delineated to identify a set of 56 structures in the brain, most of which are within the cortex. These delineations were then transformed into a common atlas space to produce a set of co-registered anatomical labels. The original MRI data were also transformed into the atlas space. The 3D volumes contained within this data set represent:

  • Intensity averages of the co-registered skull-stripped MRI volumes.
  • Probabilistic tissue maps based on automated classification of the native-space MRI into white matter, grey matter, and cerebrospinal fluid.
  • Probabilistic maps for each delineated structure.
  • Probabilistic maps for the grey matter portions of each structure.
  • Maximum likelihood maps that identify the most likely structure at each voxel in the atlas space.

Publications

Please cite this article as: Shattuck DW, Mirza M, Adisetiyo V, Hojatkashani C, Salamon G, Narr KL, Poldrack RA, Bilder RM, Toga AW, Construction of a 3D Probabilistic Atlas of Human Cortical Structures, NeuroImage (2007), doi: 10.1016/j.neuroimage.2007.09.031

Click here for publication

Monkey Atlas

A three-dimensional multimodality computerized map of the nemestrina monkey brain was created with serial sectioning and digital imaging techniques. An adult female Macaca nemestrina (pigtail macaque) weighing 7.2 kg was used in constructing this atlas. CT, PET, MRI and Cryo data were gathered on the subject. Whole head and brain only 3D image sets were created from the specimen blockface. The CT, PET, and MRI data sets were reconstructed into a digital volume and co-registered (CT and MRI by landmark matching and fiducials, PET by an automated ratio method) to the cryosection volume. The CT, PET, and MRI data sets were reconstructed into a digital volume and co-registered (CT and MRI by landmark matching and fiducials, PET by an automated ratio method) to the cryosection volume.

Mouse Atlas

The MAP atlas is comprised of a magnetic resonance microscopy (MRM) and an anatomic delineation volume, and a set of software tools for visualization, navigation, and registration of image volumes. The MRM volume is a diffusion-weighted image (DWI) acquired over several hours in a high-field magnet. DWI volumes show a great deal of anatomical detail and good contrast between gray and white matter. Anatomic delineations serve to help orient the user with graphical representations highlighting important anatomical detail and providing a standard description and nomenclature of the region of interest. The anatomical parcellation is based on Brain Maps (Swanson, 1998) with modifications to adapt it to mouse neuroanatomy. The delineations are overlaid upon the MRM volume for visualization.

Visit the MAP web site

Mouse Magnetic Resonance Microscopy Atlas

An atlas based on a magnetic resonance microscopy (MRM) image diffusion-weighted in the Z-direction acquired from a normal, 100-day old male C57BL/6J mouse. The atlas is comprised of a diffusion-weighted image volume, a label volume, a mask volume, and a label index.

Mouse Minimum Deformation Atlas

This atlas is constructed from T2-weighted magnetic resonance microscopy (MRM) images acquired from 11 normal female C57BL/6J mice and is the default atlas for the Mouse BIRN Atlasing Toolkit (MBAT).

Neonatal (P0) Mouse Nissl Brain Atlas

A full-color Nissl-stained volume with resolution of 6.6 x 50 x 6.6 µm3 was constructed. A sub-sampled version of this volume (with resolution of 13.3 x 50 x 13.3 µm3) was co-registered to a standard anatomical space defined by an averaged geometry of C57BL/6J P0 mouse brains. A hundred and forty-five anatomical structures were delineated based on the histological images. Anatomical relationships of delineated structures were established based on the hierarchical relations defined in the atlas of adult mouse brain so the P0 atlas can be related to the database associated with the adult atlas.

Neonatal (P0) MRI Mouse Brain Atlas

The atlas was constructed from the average of eight individual co-registered MR image volumes of C57BL/6J mouse brain collected at postnatal day 0. The atlas space defined by multi animals captured the average geometry of native anatomical space for normal mouse brains at P0, avoiding possible anatomical outliers present in one animal. The label volume based on this average brain is consisted of 13 anatomic structures, which provides a template to facilitate feature-based registration. The probability maps for individual anatomical volumes were defined by the eight -registered MR images used to construct the atlas, which serve as references to map the atlas onto test images and to label the regions of interest of test images. The probability map with a restrictive threshold provides a more accurate mapping, while that with lower threshold labels an image more completely.

The atlas is comprised of:

  1. an image volume of the average brain.
  2. an anatomical label volume ( 13 structures)based on the average brain, and anatomical label volumes ( 13 structures ) retrieved from the probability maps at threshold of 62.5%, 75%, 87.5% and 100% confidences estimated from the population used to construct the atlas ( all structures here are smaller then the sizes of their corresponding average structures ). The names of the structures can be referenced from their intensities of the label volumes with the index table provided.
  3. probability maps for each of the 13 anatomical structures. The probability values in space were encoded with intensities.
  4. sub-volumes for each anatomical structure tessellated from anatomical label volumes described in 2 and as well retrieved from the probability maps with thresholds at 50% confidences. These are volumes with binary values 1 and 0.

There are a number of different applications that can be used to view the atlas. Here is a list of a few of them:

Rat Atlas

The Rat Atlas is a three dimensional (3D) computerized map of rat brain anatomy created with digital imaging techniques. Electronic sharing and interactive use are benefits afforded by a digital format, but the foremost advantage of this 3D map is its whole brain integrated representation of rat in situ neuroanatomy.

Vervet Atlas

Cryo atlas:
Images have been registered to an MRI atlas generated from 357 male and female vervets three years of age or older and reflect the mean affine size and shape of the population, not those of the specific animal from which these images were generated. "MRI stereotaxic coordinates" are relative to the mean MRI atlas anterior commissure, which is anterior to the anterior commissure of the particular animal shown here. "This atlas" coordinates are slice numbers in the high resolution atlas shown here (0.1667 mm thick slices). "MRI atlas" coordinates are slice numbers in the corresponding MRI atlas (0.5 mm thick slices). Click on listed coordinates to modify them.
 
Average MRI atlas:
Data is from a population of 357 Chlorocebus aethiops sabaeus male and female monkeys. The images have been registered and scaled to reflect the mean affine size and shape of the population. Ten animals were with structural abnormalities were omitted in generating the intensity averaged images shown here. All animals were three years of age or older. "MRI stereotaxic coordinates" are relative to the mean MRI atlas anterior commissure. "This atlas" coordinates are slice numbers in the (0.5 mm thick slices) atlas shown here. "Brain atlas" coordinates are slice numbers in the corresponding high resolution (0.1667 mm) anatomic atlas. Please click on listed coordinates to modify them.