Detecting normal pediatric brain development with diffusional kurtosis imaging

Purpose: To characterise the pattern of change of diffusional kurtosis imaging (DKI) parameters (including kurtosis and diffusion parameters) in both white matter and gray matter in normal brain development with a large sample of subjects from term-born neonates to 14-years old children. Methods: Two hundred and eighteen normal children (136 male, 82 female) underwent conventional magnetic resonance imaging and DKI. Regions of interest (ROIs) were placed in 7 white matter areas and 4 gray matter areas. Then the DKI-derived parameters were automatically calculated, including fractional anisotropy (FA), mean diffusivity (MD), axial diffusivity (Da), radial diffusivity (Dr), mean kurtosis (MK), axial kurtosis (Ka) and radial kurtosis (Kr). The correlation between the DKI parameters and ages were analyzed using nonlinear fit, and the rate of parameter change was computed compared to the baseline value of the neonates. Results: For all ROIs in the white matter and gray matter, the FA, MK, Kr, Ka values increased with age, while the MD and Dr values decreased with age. The correlations were good to excellent, which changed rapidly within the first 2 years and relatively slowly after 2 years. The Da values in peripheral white matters and some gray matter structures (caudate nucleus and putamen) decreased with age. The amplitude of kurtosis parameters variation was greater than that of the diffusion parameters in both white matter and gray matter. Conclusions: The DKI parameters correlated well with age, and kurtosis parameters showed a potential advantage in detecting the normal brain development of children.