Deep brain stimulation (DBS) is an effective surgical procedure for the treatment of several brain disorders. However, the clinical successes of DBS hinges on several factors. Here, we describe the development of tools and methodologies in the context of thalamic DBS for essential tremor (ET) to address three key challenges: 1) accurate localization of nuclei and fiber pathways for stimulation, 2) model-based programming of high-density DBS electrode arrays (DBSA) and 3) in vivo assessment of computational DBS model predictions. We approached the first challenge through a multimodal imaging approach, utilizing high-field (7T) susceptibility-weighted imaging and diffusion-weighted imaging data. A nonlinear image deformation algorithm was used in conjunction with probabilistic fiber tractography to segment individual thalamic sub-nuclei and reconstruct their afferent fiber pathways. We addressed the second challenge by developing subject-specific computational model-based algorithms built on maximizing population activating function values within a target region using convex optimization principles. The algorithms converged within seconds and only required as many finite-element simulations as the number of electrodes on the DBSA being modeled. For the third challenge, we recorded (in two non-human primates) unit-spike data from neurons in the vicinity of chronically implanted thalamic DBSAs before, during and after high-frequency stimulation. A novel entropy-based method was developed to quantify the degree and significance of stimulation-induced changes in neuronal firing pattern. Results indicated that neurons modulated by thalamic DBS were distributed and not confined to the immediate proximity of the active electrode. For those that were modulated by DBS, their responses increasingly shifted from firing rate modulation to firing pattern modulation with increased stimulation amplitude. Additionally, strong low-pass filtering effect was observed where <4% of DBS pulses produced phase-locked spikes in cells exhibiting significant excitatory firing pattern modulation. Finally, we quantified the spatial distribution of neurons modulated by DBS by developing a novel spherical statistical framework for analysis. Together, these tools and methodologies are poised to improve our understanding of DBS mechanisms and improve the efficacy and efficiency of DBS therapy.
University of Minnesota Ph.D. dissertation. September 2016. Major: Biomedical Engineering. Advisor: Matthew Johnson. 1 computer file (PDF); xii, 181 pages.
Experimental and Model-based Approaches to Directional Thalamic Deep Brain Stimulation.
Retrieved from the University of Minnesota Digital Conservancy,
Content distributed via the University of Minnesota's Digital Conservancy may be subject to additional license and use restrictions applied by the depositor.