Finite Element Electrode and Individual Patient Modeling to Optimize Restorative Neuroengineering
Malaga, Karlo
2019
Abstract
Parkinson disease (PD) and essential tremor (ET) are the most common neurological movement disorders among adults. Deep brain stimulation (DBS) is an established surgical treatment for both conditions that involves implanting electrodes in the brain and then applying electrical stimulation. Despite the clinical effectiveness of DBS, its underlying mechanisms remain unclear. As DBS advances into a viable treatment for other conditions, it has become important to address the fundamental principles behind the procedure, specifically the spatial extent of stimulation. Furthermore, as DBS moves toward the adoption of closed-loop stimulation paradigms, an increased understanding of how neural recordings are affected by different biological factors is also key. Broadly, this work utilizes finite element electrode and individual patient modeling in an effort to help improve established procedures within Restorative Neuroengineering. The first study presents an atlas-independent, n-of-1 tissue activation modeling methodology utilizing diffusion tensor imaging to map the optimal location of subthalamic DBS for PD. The volume of tissue activated (VTA) was modeled using finite element analysis for 40 PD patients. High variability in neuroanatomy, stimulation location, and motor improvement was observed across patients, highlighting the need for n-of-1 modeling approaches. The optimal stimulation location was mapped to the dorsolateral border of the subthalamic nucleus, in the posterior half. Therapeutic VTAs spread further in the dorsal direction, providing more evidence for caudal zona incerta as an important DBS target. The second study applies the VTA modeling methodology from the previous study to thalamic DBS for ET and combines it with atlas-independent, n-of-1 thalamic segmentation to evaluate the effects of DBS on individual patients. Twenty-two ET patients were modeled. Thalami were segmented into 13 distinct subnuclei using a k-means clustering algorithm. Therapeutic and side effect-inducing VTAs were calculated. Results were compared with those obtained using an atlas-based segmentation approach. Within the thalamus, the shape and size of the VTA was highly variable across patients. Overall, n-of-1 segmentation performed better than atlas-based segmentation in explaining DBS side effects. The third study investigates the effects of gliosis and interface interactions at the electrode recording site on single-unit recording quality using a computational model incorporating impedance and neural data from Utah arrays implanted in rhesus macaques. A finite element model of a Utah array microelectrode in neural tissue was coupled with a multi-compartment neuron model to quantify the effects of encapsulation thickness, encapsulation resistivity, and interface resistivity on electrode impedance and waveform amplitude. The neural recording model was then reconciled with the in vivo data. From week 1-3, mean impedance and amplitude increased at rates of 115.8 kΩ/week and 23.1 µV/week, respectively. This initial ramp up in impedance and amplitude was consistent with increasing interface resistivity and tissue resistivity, respectively, in the model. The modeled gliosis could not match the in vivo data. However, a thin interface layer at the recording site could. Despite having a large effect on impedance, interface resistivity did not have a noticeable effect on amplitude, suggesting that gliosis does not cause an electrical problem with regard to signal quality. The significance of this work is in the n-of-1 modeling framework that it provides. As insight regarding stimulation spread in the brain increases, the techniques described can be applied to other conditions to inform novel stimulation strategies and help bridge the gap between model-based evidence and clinical practice.Subjects
Neural engineering Computational modeling Finite element method Deep brain stimulation Movement disorders Neural recording
Types
Thesis
Metadata
Show full item recordCollections
Remediation of Harmful Language
The University of Michigan Library aims to describe library materials in a way that respects the people and communities who create, use, and are represented in our collections. Report harmful or offensive language in catalog records, finding aids, or elsewhere in our collections anonymously through our metadata feedback form. More information at Remediation of Harmful Language.
Accessibility
If you are unable to use this file in its current format, please select the Contact Us link and we can modify it to make it more accessible to you.