A large number of models have been developed to describe the diffusion MRI signal as a sum of different neural compartments. However, development and optimization of these multicompartment models has largely focused on the brain. In this work, we apply and compare a combinatorially large number of biophysical models (N=480) in the in vivo human spinal cord, evaluating their ability to fit the signal and also predict unseen signal. We find that certain combinations of constraints and compartments better model the signal in the cervical spinal cord, and we give recommendations for future modeling of this structure with clinical acquisitions.
This abstract and the presentation materials are available to members only; a login is required.