There is much interest in using magnetic resonance diffusion imaging to provide information on anatomical connectivity in the brain by measuring the diffusion of water in white matter tracts. Among the measures, the most commonly derived from diffusion data is fractional anisotropy (FA), which quantifies local tract directionality and integrity. Many multi-subject imaging studies are using FA images to localize brain changes related to development, degeneration and disease. In a recent paper, we presented a new approach, tract-based spatial statistics (TBSS), which aims to solve crucial issues of cross-subject data alignment, allowing localized cross-subject statistical analysis. This works by transforming the data from the centers of the tracts that are consistent across a study's subjects into a common space. In this protocol, we describe the MRI data acquisition and analysis protocols required for TBSS studies of localized change in brain connectivity across multiple subjects.

NOTE: In the version of this article originally published online, the URL given in EQUIPMENT SETUP, under “Computing equipment”, should have been http://www.fmrib.ox.ac.uk/fsl. In the first line of Box 2, “head motion” should have read “head motion effects”. In the legend to Figure 1, “and a ˙ b value" should have been “and a b value” and "an signal-to-noise ratio" should have been "a signal-to-noise ratio". These errors have been corrected in all versions of the article.