Voxel-Based Morphometry

The human brain is in a state of constant change and adaptation. This may be driven either by normal developmental or aging processes or by the effects of learning, training, and new occurrences in daily life. In addition to these aforementioned changes, more systematic influences such as gender, disease, and genes affect the brain’s structure. Using magnetic resonance imaging, brain changes and differences can be measured noninvasively and in vivo, making them particularly interesting for both basic research and clinical research. The easiest way to assess brain changes (or group differences) is to measure whole-brain volume. However, assessing the volume of the entire brain is rather unspecific. So-called region-of-interest (ROI) analyses are more sensitive to local changes than are whole-brain assessments but are also subject to several limitations. For example, if one specific region is measured, other brain structures are ignored, and possible effects remain undetected elsewhere in the brain. Moreover, ROIs are usually created based on individual protocols and depend on raterspecific ‘judgment calls,’ thus requiring a clearly definable and unambiguous structure. For large parts of the brain, however, it may be difficult to precisely define (or identify) unambiguous boundaries. Finally, if an ROI is only partially different, this will lower the sensitivity to detect any effects in this region. This is where voxel-based morphometry (VBM) comes into play, as VBM allows for the examination of brain changes and/or group differences across the entire brain with a high regional specificity (i.e., voxel by voxel), without requiring the a priori definition of particular ROIs (Ashburner & Friston, 2000, 2001, 2007).