Central nervous system disorders produce the undesired, approximately rhythmic movement of body parts known as pathological tremor. This undesired motion inhibits the patient's ability to perform tasks of daily living and participate in society. Typical treatments are medications and deep brain stimulation surgery, both of which include risks, side effects, and varying efficacy. Since the pathophysiology of tremor is not well understood, empirical investigation drives tremor treatment development. This dissertation explores tremor from a mechanical systems perspective to work towards theory-driven treatment design. The primary negative outcome of pathological tremor is the undesired movement of body parts: mechanically suppressing this motion provides effective tremor treatment by restoring limb function. Unlike typical treatments, the mechanisms for mechanical tremor suppression are well understood: applying joint torques that oppose tremor-producing muscular torques will reduce tremor irrespective of central nervous system pathophysiology. However, a tremor suppression system must also consider voluntary movements. For example, mechanically constraining the arm in a rigid cast eliminates tremor motion, but also eliminates the ability to produce voluntary motions. Indeed, passive mechanical systems typically reduce tremor and voluntary motions equally due to the close proximity of their frequency content. Thus, mechanical tremor suppression requires active actuation to reduce tremor with minimal influence on voluntary motion. However, typical engineering actuators are rigid and bulky, preventing clinical implementations. This dissertation explores dielectric elastomers as tremor suppression actuators to improve clinical implementation potential of mechanical tremor suppression. Dielectric elastomers are often called "artificial muscles" due to their similar mechanical properties as human muscle; these similarities may enable relatively soft, low-profile implementations. The primary drawback of dielectric elastomers is their relatively low actuation levels compared to typical actuators. This research develops a tremor-active approach to dielectric elastomer-based tremor suppression. In a tremor-active approach, the actuators only actuate to oppose tremor, while the human motor system must overcome the passive actuator dynamics. This approach leverages the low mechanical impedance of dielectric elastomers to overcome their low actuation levels. Simulations with recorded tremor datasets demonstrate excellent and robust tremor suppression performance. Benchtop experiments validate the control approach on a scaled system. Since dielectric elastomers are not yet commercially available, this research quantifies the necessary dielectric elastomer parameters to enable clinical implementations and evaluates the potential of manufacturing approaches in the literature to achieve these parameters. Overall, tremor-active control using dielectric elastomers represents a promising alternative to medications and surgery. Such a system may achieve comparable tremor reduction as medications and deep brain stimulation with minimal risks and greater efficacy, but at the cost of increased patient effort to produce voluntary motions. Parallel advances in scaled dielectric elastomer manufacturing processes and high-voltage power electronics will enable consumer implementations. In addition to tremor suppression, this dissertation investigates the mechanisms of central nervous system tremor generation from a control systems perspective. This research investigates a delay-based model for parkinsonian tremor. Besides tremor, Parkinson's disease generally inhibits movement, with typical symptoms including rigidity, bradykinesia, and increased reaction times. This fact raises the question as to how the same disease produces excessive movement (tremor) despite characteristically inhibiting movement. One possible answer is that excessive central nervous system inhibition produces unaccounted feedback delays that cause instability. This dissertation develops an optimal control model of human motor control with an unaccounted delay between the state estimator and controller. This delay represents the increased inhibition projected from the basal ganglia to the thalamus, delaying signals traveling from the cerebellum (estimator) to the primary motor cortex (controller). Model simulations show increased delays decrease tremor frequency and increase tremor amplitude, consistent with the evolution of tremor as the disease progresses. Simulations that incorporate tremor resetting and random variation in control saturation produce simulated tremor with similar characteristics as recorded tremor. Delay-induced tremor explains the effectiveness of deep brain stimulation in both the thalamus and basal ganglia since both regions contribute to the presence of feedback delay. Clinical evaluation of mechanical tremor suppression may provide clinical evidence for delay-induced tremor: unlike state-independent tremor, suppression of delay-induced tremor increases tremor frequency. Altogether, establishing the mechanisms for tremor generation will facilitate pathways towards improved treatments and cure development.
If this is your thesis or dissertation, and want to learn how to access it or for more information about readership statistics, contact us at STARS@ucf.edu
Kauffman, Jeffrey L.
Doctor of Philosophy (Ph.D.)
College of Engineering and Computer Science
Mechanical and Aerospace Engineering
Length of Campus-only Access
Doctoral Dissertation (Open Access)
Kelley, Christopher, "Pathological Tremor as a Mechanical System: Modeling and Control of Artificial Muscle-Based Tremor Suppression" (2020). Electronic Theses and Dissertations, 2020-. 70.