The Effects of Long-Term Application of Transcranial Direct Current Stimulation on Transfer of Motor Learning in Parkinson’s Disease

s Brain Stimulation 14 (2021) 1589e1707 We use our previously developed stochastic differential equation approach to model segmentation errors as a head shape boundary uncertainty. This approach treats the output of the segmentation as the nominal tissue boundary and it generates realizations of the tissue boundary by adding random perturbations along the normal direction for each point. For each tissue boundary realization, we then use an in-house boundary element method to perform a forward propagation analysis and quantify the impact of tissue boundary uncertainties on the induced E-field. Our results show that TMS induced cortical E-fields are largely insensitive to scalp, skull, and white matter compartment segmentation errors. In contrast, the E-fields are highly sensitive to CSF compartment segmentation error. Furthermore, the skull/CSF boundary uncertainty results in highest uncertainty in the E-fields induced on the Gyral Crowns, whereas CSF/GM boundary uncertainty results in uncertainty in the Sulci. Averaging the E-field over a region lowers uncertainty. The accuracy of current E-field simulations is limited by the accuracy of CSF compartment discretization. Furthermore, the averaging an E-field over a region provides a dose that is less sensitive to head segmentation errors.