DTI analysis methods: Voxel-based analysis

Abstract

Voxel-based analysis (VBA) of diffusion tensor imaging (DTI) data permits the investigation of voxel-wise differences or changes in DTI metrics in every voxel of a brain dataset. It is applied primarily in the exploratory analysis of hypothesized group-level alterations in DTI parameters, as it does not require prior knowledge of where in the brain such changes may occur. Whilst VBA is a widely used, powerful preclinical research tool, there are a number of methodological issues that should be considered when applying the technique to study (pre)clinical populations. This chapter reviews the component steps of a typical VBA study pipeline and includes a comprehensive introduction to image registration, DTI template/atlas selection, smoothing, and statistical analysis. The popular tract-based spatial (TBSS) technique is introduced and contrasted with traditional VBA approaches. At each stage, guidance on optimizing parameter settings is presented along with the pros and cons of different methods to assist the reader in choosing the best approach for their application.