Elsevier

NeuroImage

Volume 62, Issue 2, 15 August 2012, Pages 911-922
NeuroImage

Review
Brain templates and atlases

https://doi.org/10.1016/j.neuroimage.2012.01.024Get rights and content

Abstract

The core concept within the field of brain mapping is the use of a standardized, or “stereotaxic”, 3D coordinate frame for data analysis and reporting of findings from neuroimaging experiments. This simple construct allows brain researchers to combine data from many subjects such that group-averaged signals, be they structural or functional, can be detected above the background noise that would swamp subtle signals from any single subject. Where the signal is robust enough to be detected in individuals, it allows for the exploration of inter-individual variance in the location of that signal. From a larger perspective, it provides a powerful medium for comparison and/or combination of brain mapping findings from different imaging modalities and laboratories around the world. Finally, it provides a framework for the creation of large-scale neuroimaging databases or “atlases” that capture the population mean and variance in anatomical or physiological metrics as a function of age or disease.

However, while the above benefits are not in question at first order, there are a number of conceptual and practical challenges that introduce second-order incompatibilities among experimental data. Stereotaxic mapping requires two basic components: (i) the specification of the 3D stereotaxic coordinate space, and (ii) a mapping function that transforms a 3D brain image from “native” space, i.e. the coordinate frame of the scanner at data acquisition, to that stereotaxic space. The first component is usually expressed by the choice of a representative 3D MR image that serves as target “template” or atlas. The native image is re-sampled from native to stereotaxic space under the mapping function that may have few or many degrees of freedom, depending upon the experimental design. The optimal choice of atlas template and mapping function depend upon considerations of age, gender, hemispheric asymmetry, anatomical correspondence, spatial normalization methodology and disease-specificity. Accounting, or not, for these various factors in defining stereotaxic space has created the specter of an ever-expanding set of atlases, customized for a particular experiment, that are mutually incompatible.

These difficulties continue to plague the brain mapping field. This review article summarizes the evolution of stereotaxic space in term of the basic principles and associated conceptual challenges, the creation of population atlases and the future trends that can be expected in atlas evolution.

Introduction

The creation and application of atlases for brain research are now widespread (see Cabezas et al., 2011 for a comprehensive review). Here we focus on the historical evolution of those atlases and strategies that have particular relevance for the fMRI community. The core concept within the field of brain mapping is the use of a standardized 3D coordinate frame for the analysis and reporting of neuroimaging experiments. This apparently trivial device has had a profound impact on the study of functional neuroanatomy and the practice of neuroscience research in general. The early pioneering works of Brodmann, 1909, Brodmann, 1914, Brodmann, 1960, Flechsig (1920), von Economo and Koskinas (1925), Vogt and Vogt (1919) and Sarkisov et al. (1955) to map the cyto- and myelo-architectural landscape of the human cortex were based on painstaking visual inspection and characterization of a few observable cellular properties (packing density, cell shape, laminar organization). They were typically carried out on single brains and cortical parcellation schemes were largely restricted to the gyral surface, ignoring the two-thirds of cortical area buried in the sulcal folds (Zilles and Amunts, 2010). Neuroimaging techniques evolved over the last 20 years have allowed neuroscientists to re-visit the issue of mapping the human brain, such that a modern brain atlas is now expressed as a digital database that can capture the spatio-temporal distribution of a multitude of physiological and anatomical metrics. Furthermore, it allows for a quantitative characterization of the (i) normal variability in those metrics across a population, (ii) the relationship between those metrics and behavioral performance and (iii) the detection of subtle changes associated with disease, genotype, gender or demographics.

Section snippets

Whole brain adult atlases

The origins of modern brain mapping lie with the seminal work of Jean Talairach (Talairach et al., 1967), who developed a 3D coordinate space to assist deep-brain surgical techniques. This was later updated by Talairach and Tournoux (1988) as a printed atlas for guidance of deep-brain stereotactic procedures. The earliest application of Talairach space for brain mapping was by Fox et al. (1985) who used it to map the 3D coordinates of activation foci from PET experiments in different

Diffeomorphism and anatomical variability

The 3D spatial mapping function between individual brains and a template brain may take many forms: from a simple, linear transformation (e.g. Collins et al., 1994, Woods et al., 1993) to more sophisticated non-linear approaches. While the linear approaches have the advantage of simplicity, and therefore easier comparability across studies from different laboratories, they exhibit poorer anatomical correspondence than higher-dimensional approaches. Fig. 3a summarizes schematically the

Current and future trends

The progression of brain atlases over the last 20 years from printed reference textbooks to digital, multivariate, multidimensional, statistical maps has been quite staggering. There appears to be no end in sight as computational tools become ever-more powerful and accessible, and data sharing, a common practice of the neuroimaging community. The modern brain atlas is now an engine for meta-analysis of information from all corners of neuroscience. We can expect the next decade to bring forth a

Conclusion

The combination of the traditional atlas metrics of classical neuroanatomy with maps of gene expression and connectivity, potentially with a temporal component, will offer entirely novel perspectives on brain organization. Capturing the population variance of these multivariate attribute maps in the form of probabilistic atlases, amenable to statistical analysis, will revolutionize our understanding of functional neuroanatomy and the changes associated with development, aging, learning and

References (207)

  • F.L. Bookstein

    Biometrics, biomathematics and the morphometric synthesis

    Bull. Math. Biol.

    (1996)
  • F.L. Bookstein

    Landmark methods for forms without landmarks: morphometrics of group differences in outline shape

    Med. Image Anal.

    (1997)
  • E.D. Burgund et al.

    The feasibility of a common stereotactic space for children and adults in fMRI studies of development

    NeuroImage

    (2002)
  • M. Cabezas et al.

    A review of atlas-based segmentation for magnetic resonance brain images

    Comput. Methods Programs Biomed.

    (2011)
  • P.S. Carmack et al.

    Improved agreement between Talairach and MNI coordinate spaces in deep brain regions

    NeuroImage

    (2004)
  • S. Caspers et al.

    The human inferior parietal cortex: cytoarchitectonic parcellation and interindividual variability

    NeuroImage

    (2006)
  • M. Catani et al.

    A diffusion tensor imaging tractography atlas for virtual in vivo dissections

    Cortex

    (2008)
  • A. Charil et al.

    Statistical mapping analysis of lesion location and neurological disability in multiple sclerosis: application to 452 patient data sets

    NeuroImage

    (2003)
  • A. Charil et al.

    Focal cortical atrophy in multiple sclerosis: relation to lesion load and disability

    NeuroImage

    (2007)
  • W. Chau et al.

    The Talairach coordinate of a point in the MNI space: how to interpret it

    NeuroImage

    (2005)
  • M.K. Chung et al.

    A unified statistical approach to deformation-based morphometry

    NeuroImage

    (2001)
  • R.W. Cox

    AFNI: software for analysis and visualization of functional magnetic resonance neuroimages

    Comput. Biomed. Res.

    (1996)
  • B. Crespo-Facorro et al.

    Human frontal cortex: an MRI-based parcellation method

    NeuroImage

    (1999)
  • A.M. Dale et al.

    Cortical surface-based analysis. I. Segmentation and surface reconstruction

    NeuroImage

    (1999)
  • C. Davatzikos et al.

    Voxel-based morphometry using RAVENS maps: methods and validation using simulated longitudinal atrophy

    NeuroImage

    (2001)
  • R.S. Desikan et al.

    An automated labeling system for subdividing the human cerebral cortex on MRI scans into gyral based regions of interest

    NeuroImage

    (2006)
  • C. Destrieux et al.

    Automatic parcellation of human cortical gyri and sulci using standard anatomical nomenclature

    NeuroImage

    (2010)
  • J. Diedrichsen

    A spatially unbiased atlas template of the human cerebellum

    NeuroImage

    (2006)
  • J. Diedrichsen et al.

    A probabilistic MR atlas of the human cerebellum

    NeuroImage

    (2009)
  • J. Diedrichsen et al.

    Imaging the deep cerebellar nuclei: a probabilistic atlas and normalization procedure

    NeuroImage

    (2011)
  • S.B. Eickhoff et al.

    A new SPM toolbox for combining probabilistic cytoarchitectonic maps and functional imaging data

    NeuroImage

    (2005)
  • S.B. Eickhoff et al.

    Assignment of functional activations to probabilistic cytoarchitectonic areas revisited

    NeuroImage

    (2007)
  • S.B. Eickhoff et al.

    Activation likelihood estimation meta-analysis revisited

    Neuroimage

    (2012)
  • A.C. Evans

    NIH MRI study of normal brain development

    NeuroImage

    (2006)
  • B. Fischl et al.

    Cortical surface-based analysis. II: inflation, flattening, and a surface-based coordinate system

    NeuroImage

    (1999)
  • B. Fischl et al.

    Whole brain segmentation: automated labeling of neuroanatomical structures in the human brain

    Neuron

    (2002)
  • V. Fonov et al.

    Unbiased average age-appropriate atlas for pediatric studies

    NeuroImage

    (2011)
  • P.T. Fox et al.

    Beyond the single study: function–location meta-analysis in cognitive neuroimaging

    Curr. Opin. Neurobiol.

    (1998)
  • S. Geyer et al.

    Areas 3a, 3b, and 1 of human primary somatosensory cortex. Part 2 Spatial normalization to standard anatomical space

    NeuroImage

    (2000)
  • S.S. Ghosh et al.

    Evaluating the validity of volume-based and surface-based brain image registration for developmental cognitive neuroscience studies in children 4-to-11 years of age

    NeuroImage

    (2010)
  • C.D. Good et al.

    Cerebral asymmetry and the effects of sex and handedness on brain structure: a voxel-based morphometric analysis of 465 normal adult human brains

    NeuroImage

    (2001)
  • C. Grefkes et al.

    Human somatosensory area 2: observer-independent cytoarchitectonic mapping, interindividual variability, and population map

    NeuroImage

    (2001)
  • K. Hua et al.

    Tract probability maps in stereotaxic spaces: analyses of white matter anatomy and tract-specific quantification

    NeuroImage

    (2008)
  • M. Altaye et al.

    Infant brain probability templates for MRI segmentation and normalization

    NeuroImage

    (2008)
  • K. Amunts et al.

    Broca's region revisited: cytoarchitecture and intersubject variability

    J. Comp. Neurol.

    (1999)
  • J. Ashburner et al.

    Nonlinear spatial normalization using basis functions

    Hum. Brain Mapp.

    (1999)
  • J. Ashburner et al.

    Multivariate models of inter-subject anatomical variability

    NeuroImage

    (2010)
  • B. Aubert-Broche et al.

    Twenty new digital brain phantoms for creation of validation image databases

    IEEE Trans. Image Med.

    (2006)
  • G. Auzias et al.

    Diffeomorphic brain registration under exhaustive sulcal constraints

    IEEE Trans. Med. Imaging

    (2011)
  • S. Baillet et al.

    Electromagnetic brain mapping

    IEEE Signal Process. Mag.

    (2001)
  • Cited by (0)

    View full text