Most model-based diffusion metrics (AxCaliber metrics) have been shown to be less stable and more biased on clinical systems due to the limited gradient strength (40-80mT/m versus >300mT/m on preclinical scanners). In this work we wanted to (i) find the best AxCaliber protocol at 80mT/m and (ii) quantify the bias in the estimated metrics. For these aims, we first optimized an 80mT/m AxCaliber protocol using simulations, then compared experimentally (on an ex vivo cat spinal cord) a 600mT/m protocol versus the optimized 80mT/m protocol. Using the 600mT/m maps as a ground truth, our results show that even though axon diameter cannot be estimated robustly, the fraction of restricted water can be measured accurately (<3% error) and precisely (r2 >0.76) on clinical systems. The short duration of the optimized protocol opens the way to the use of reliable model-based diffusion MRI metrics on a clinical system, metrics that would be particularly useful to measure the degree of myelination through the fiber g-ratio.
This abstract and the presentation materials are available to members only; a login is required.