Inherently low signal-to-noise ratio in diffusion-MRI results in high uncertainty in the estimated underlying fibre orientation distributions. True bootstrapping and model- based residual bootstrapping approaches have been proposed to estimate the uncertainty in the estimated peak directions. Here we compare the two methods using simulations and observe that more information can be extracted by looking at the variation in the entire shape of the FOD, as opposed to just the peak direction. Results also suggest that, in this case, residual bootstrapping might not provide a representative estimation of directional uncertainty due to non-negligible model bias.
This abstract and the presentation materials are available to members only; a login is required.