Global sensitivity analysis of the probabilistic volume of tissue activated in a volume conductor model for deep brain stimulation

To date, various computational models exist to gain insight into the mechanisms of action of deep brain stimulation (DBS) as well as to predict the volume of tissue activated (VTA). The parameters of these models are subject to uncertainty, and knowledge about the sensitivity of the VTA on this uncertainty is scarce. The aim of this study was to perform a global sensitivity analysis of the probabilistic VTA on the uncertainty in the electrical properties of brain tissue as well as on the parameters of the electrode-tissue-interface, which is present under voltage controlled stimulation. Global sensitivity analysis was carried out using the polynomial chaos technique, which is based on polynomial expansions in random space. The results showed a major influence of the electrical conductivity of brain tissue, which suggests that the uncertainty of the tissue parameters is a crucial factor in the computational modeling of DBS.