Voxel-based morphometry is a computational approach to

neuroanatomy Neuroanatomy is the study of the structure and organization of the nervous system. In contrast to animals with radial symmetry, whose nervous system consists of a distributed network of cells, animals with bilateral symmetry have segregated, defin ...

that measures differences in local concentrations of brain tissue, through a voxel-wise comparison of multiple brain images. In traditional morphometry, volume of the whole brain or its subparts is measured by drawing regions of interest (ROIs) on images from brain scanning and calculating the

volume Volume is a measure of occupied three-dimensional space. It is often quantified numerically using SI derived units (such as the cubic metre and litre) or by various imperial or US customary units (such as the gallon, quart, cubic inch). Th ...

enclosed. However, this is time consuming and can only provide measures of rather large areas. Smaller differences in volume may be overlooked. The value of VBM is that it allows for comprehensive measurement of differences, not just in specific structures, but throughout the entire brain. VBM registers every brain to a template, which gets rid of most of the large differences in brain anatomy among people. Then the brain images are smoothed so that each

voxel In 3D computer graphics, a voxel represents a value on a regular grid in three-dimensional space. As with pixels in a 2D bitmap, voxels themselves do not typically have their position (i.e. coordinates) explicitly encoded with their values. ...

represents the average of itself and its neighbors. Finally, the image volume is compared across brains at every voxel. However, VBM can be sensitive to various artifacts, which include misalignment of brain structures, misclassification of tissue types, differences in folding patterns and in cortical thickness. All these may confound the statistical analysis and either decrease the sensitivity to true volumetric effects, or increase the chance of false positives. For the cerebral cortex, it has been shown that volume differences identified with VBM may reflect mostly differences in surface area of the cortex, than in cortical thickness.

History

Over the past two decades, hundreds of studies have shed light on the neuroanatomical structural correlates of neurological and psychiatric disorders. Many of these studies were performed using voxel-based morphometry (VBM), a whole-brain technique for characterizing between groups' regional volume and tissue concentration differences from structural magnetic resonance imaging (MRI) scans. One of the first VBM studies and one that came to attention in mainstream media was a study on the

hippocampus The hippocampus (via Latin from Greek , ' seahorse') is a major component of the brain of humans and other vertebrates. Humans and other mammals have two hippocampi, one in each side of the brain. The hippocampus is part of the limbic system, ...

brain structure of London taxicab drivers. The VBM analysis showed the back part of the posterior hippocampus was on average larger in the taxi drivers compared to control subjects while the anterior hippocampus was smaller. London taxi drivers need good spatial navigational skills and scientists have usually associated hippocampus with this particular skill. Another famous VBM paper was a study on the effect of age on gray and white matter and CSF of 465 normal adults. The VBM analysis showed global gray matter was decreased linearly with age, especially for men, whereas global white matter did not decline with age. A key description of the methodology of voxel-based morphometry is ''Voxel-Based Morphometry—The Methods''—one of the most cited articles in the journal ''

NeuroImage ''NeuroImage'' is a peer-reviewed scientific journal covering research on neuroimaging, including functional neuroimaging and functional human brain mapping. The current Editor in Chief is Michael Breakspear. Abstracts from the annual meetin ...

''. The usual approach for statistical analysis is mass-univariate (analysis of each voxel separately), but

pattern recognition Pattern recognition is the automated recognition of patterns and regularities in data. It has applications in statistical data analysis, signal processing, image analysis, information retrieval, bioinformatics, data compression, computer graphic ...

may also be used, e.g., for classifying patients from healthy.

For brain asymmetry

Usually VBM is performed for examining differences across subjects, but it may also be used to examine neuroanatomical differences between hemispheres detecting brain asymmetry. A technical procedure for such an investigation may use the following steps: # Construction of a study-specific brain image template with a balanced set of left and right handed and males and females. # Construction of

white White is the lightness, lightest color and is achromatic (having no hue). It is the color of objects such as snow, chalk, and milk, and is the opposite of black. White objects fully diffuse reflection, reflect and scattering, scatter all the ...

and grey matter templates from segmentation. # Construction of symmetric grey and white matter templates by averaging right and left

cerebral hemisphere The vertebrate cerebrum (brain) is formed by two cerebral hemispheres that are separated by a groove, the longitudinal fissure. The brain can thus be described as being divided into left and right cerebral hemispheres. Each of these hemispheres ...

s. # Segmentation and extraction of brain image, e.g., removal of scalp tissue in the image. #

Spatial normalization In neuroimaging, spatial normalization is an image processing step, more specifically an image registration method. Human brains differ in size and shape, and one goal of spatial normalization is to deform human brain scans so one location in ...

to the symmetric templates # Correction for volume change (applying a Jacobian determinant) # Spatial smoothing (intensity in each voxel is a local weighted average generally expressed as GM, WM, CSF concentration). # Actual statistical analysis by the

general linear model The general linear model or general multivariate regression model is a compact way of simultaneously writing several multiple linear regression models. In that sense it is not a separate statistical linear model. The various multiple linear regr ...

, i.e., statistical parametric mapping. The outcome of these steps is a statistical parametric map, highlighting all voxels of the brain where intensities (volume or GM concentration depending on whether the modulation step has been applied or not) in a group images are significantly lower/higher than those in the other group under investigation.

Compared to the region of interest approach

Before the advent of VBM, the manual delineation of region of interest was the gold standard for measuring the volume of brain structures. However, compared to the region of interest approach, VBM presents a large number of advantages that explain its wide popularity within the neuroimaging community. Indeed, it is an automated and relatively easy-to–use, time-efficient, whole-brain tool that could detect the focal microstructural differences in brain anatomy in vivo between groups of individuals without requiring any a priori decision concerning which structure to evaluate. Moreover, VBM exhibits comparable accuracy to manual volumetry. Indeed, several studies have shown good correspondence between the two techniques, providing confidence in the biological validity of the VBM approach.

References

{{reflist, 30em

History

For brain asymmetry

Compared to the region of interest approach

See also

References

Further reading