Fig. 7.

The estimated amount of change in the summary measurements as a result of a unit change in each parameter (${\mu }_{\widehat{\Delta v}}$) for a sample white matter and grey matter voxel. The error bars show the estimated standard deviation of change. colours correspond to parameters and columns indicate summary measurements. Due to differences in the baseline, each voxel can have a different change vector for the same parameter change. This added degree of freedom can model the variability of parameters (e.g. diffusivities) across the brain, which is not considered in constrained models; e.g. NODDI.