Performance analysis of single-site and multiple-site deep brain stimulation in basal ganglia for Parkinson's disease

Deep brain stimulation is an effective surgical treatment for advanced Parkinson's disease in patients whose symptoms are not well controlled by drugs. The subthalamic nucleus (STN), the globus pallidus pars externa (GPe), and the globus pallidus pars interna (GPi) are the targets for DBS. Despite over a decade of the practical clinical experience, the selections of DBS parameters are still challenging. DBS affects the local cells within the target site as well as the passing axons within neighboring regions. A computational model of basal ganglia-thalamic network is formed to analyze the performance of DBS at STN, GPe, and GPi in terms of error index. In addition to single-site stimulation, multiple-site stimulation is also analyzed for DBS of STN-GPe and STN-GPi in terms of error index. The goal of this work is to use computational modeling to analyze the effect of DBS of single-site stimulation (STN, GPe, or GPi) and multiple-site stimulation (STN-GPe or STN-GPi) with respect to stimulation amplitude and frequency.