Diffusion kurtosis imaging with tract-based spatial statistics reveals white matter alterations in preschool children

Abstract

Diffusion kurtosis imaging (DKI), an extension of diffusion tensor imaging (DTI), provides a practical method to describe non-Gaussian water diffusion in neural tissues. The sensitivity of DKI to detect the subtle changes in several chosen brain structures has been studied. However, intuitive and holistic methods to validate the merits of DKI remain to be explored. In this paper, tract-based spatial statistics (TBSS) was used to demonstrate white matter alterations in both DKI and DTI parameters in preschool children (1-6 years; n=10). Correlation analysis was also performed in multiple regions of interest (ROIs). Fractional anisotropy, mean kurtosis, axial kurtosis and radial kurtosis increased with age, while mean diffusivity and radial diffusivity decreased significantly with age. Fractional anisotropy of kurtosis and axial diffusivity were found to be less sensitive to the changes with age. These preliminary findings indicated that TBSS could be used to detect subtle changes of DKI parameters on the white matter tract. Kurtosis parameters, except fractional anisotropy of kurtosis, demonstrated higher sensitivity than DTI parameters. TBSS may be a convenient method to yield higher sensitivity of DKI.