Mathematical models of pathological oscillatory activity and effects of deep brain stimulation in movement disorders
<p>Essential tremor and Parkinson’s disease are the most common movement disorders affecting patients globally. When pharmacological treatment options have been exhausted, continuous, high-frequency deep brain stimulation (DBS) is an effective therapy option. Despite its success, DBS is limite...
| Autor Principal: | |
|---|---|
| Outros autores: | |
| Formato: | Thesis |
| Publicado: |
2020
|
| Subjects: |
| _version_ | 1826315981012598784 |
|---|---|
| author | Duchet, B |
| author2 | Bick, C |
| author_facet | Bick, C Duchet, B |
| author_sort | Duchet, B |
| collection | OXFORD |
| description | <p>Essential tremor and Parkinson’s disease are the most common movement disorders affecting patients globally. When pharmacological treatment options have been exhausted, continuous, high-frequency deep brain stimulation (DBS) is an effective therapy option. Despite its success, DBS is limited by side effects such as speech and motor impairment, as well as impulsivity. Closed-loop DBS could address these limitations by stimulating in a way that is more parsimonious and by targeting the pathology more selectively to spare physiological processing. In closed-loop DBS, pathological activity is recorded, and stimulation is delivered based on the current state and knowledge of its association with symptoms. Experimental closed-loop strategies include phase-locked DBS in essential tremor and adaptive DBS in Parkinson’s disease. In phase-locked DBS, a precise understanding of the phase-dependence of the response to stimulation is difficult to attain experimentally because of patient fatigue. In Parkinson’s disease, refining adaptive DBS likely requires a better understanding of the dynamics of bursts in the recorded beta band activity (13-35 Hz), which have been implicated in motor symptoms. To inform closed-loop DBS strategies, we use patient data as a starting point to formulate and constrain mathematical models of pathological oscillatory activity and of the effects of DBS. In Chapter 2, we model the effects of phase-locked stimulation in essential tremor patient data. Models fitted to patient data can be used to optimise DBS in silico, and provide insights into the pathophysiology of the disease. In Chapter 3, we relate changes in beta oscillation temporal patterning between medication states in Parkinson’s disease to changes in dynamical properties of the system generating beta bursts. In particular, we show that the system generating beta oscillations is more non-linear in the pathological state, which has implications for closed-loop DBS. In Chapter 4, we compare two definitions of amplitude to measure the effects of phase-locked DBS, and to optimise stimulation in models. We demonstrate that a dynamical measure of amplitude, isostable amplitude, can in some cases be more beneficial to optimise stimulation than envelope amplitude. Together, this thesis sheds light on key aspects of Parkinson’s disease and essential tremor, and contributes to advancing closed-loop DBS.</p> |
| first_indexed | 2024-03-06T20:02:16Z |
| format | Thesis |
| id | oxford-uuid:27b54d41-07a7-4781-80fd-4a384c02177f |
| institution | University of Oxford |
| last_indexed | 2024-12-09T03:35:56Z |
| publishDate | 2020 |
| record_format | dspace |
| spelling | oxford-uuid:27b54d41-07a7-4781-80fd-4a384c02177f2024-12-01T19:31:22ZMathematical models of pathological oscillatory activity and effects of deep brain stimulation in movement disordersThesishttp://purl.org/coar/resource_type/c_db06uuid:27b54d41-07a7-4781-80fd-4a384c02177fComputational neuroscienceNeurologyMathematical neuroscienceHyrax Deposit2020Duchet, BBick, CBogacz, R<p>Essential tremor and Parkinson’s disease are the most common movement disorders affecting patients globally. When pharmacological treatment options have been exhausted, continuous, high-frequency deep brain stimulation (DBS) is an effective therapy option. Despite its success, DBS is limited by side effects such as speech and motor impairment, as well as impulsivity. Closed-loop DBS could address these limitations by stimulating in a way that is more parsimonious and by targeting the pathology more selectively to spare physiological processing. In closed-loop DBS, pathological activity is recorded, and stimulation is delivered based on the current state and knowledge of its association with symptoms. Experimental closed-loop strategies include phase-locked DBS in essential tremor and adaptive DBS in Parkinson’s disease. In phase-locked DBS, a precise understanding of the phase-dependence of the response to stimulation is difficult to attain experimentally because of patient fatigue. In Parkinson’s disease, refining adaptive DBS likely requires a better understanding of the dynamics of bursts in the recorded beta band activity (13-35 Hz), which have been implicated in motor symptoms. To inform closed-loop DBS strategies, we use patient data as a starting point to formulate and constrain mathematical models of pathological oscillatory activity and of the effects of DBS. In Chapter 2, we model the effects of phase-locked stimulation in essential tremor patient data. Models fitted to patient data can be used to optimise DBS in silico, and provide insights into the pathophysiology of the disease. In Chapter 3, we relate changes in beta oscillation temporal patterning between medication states in Parkinson’s disease to changes in dynamical properties of the system generating beta bursts. In particular, we show that the system generating beta oscillations is more non-linear in the pathological state, which has implications for closed-loop DBS. In Chapter 4, we compare two definitions of amplitude to measure the effects of phase-locked DBS, and to optimise stimulation in models. We demonstrate that a dynamical measure of amplitude, isostable amplitude, can in some cases be more beneficial to optimise stimulation than envelope amplitude. Together, this thesis sheds light on key aspects of Parkinson’s disease and essential tremor, and contributes to advancing closed-loop DBS.</p> |
| spellingShingle | Computational neuroscience Neurology Mathematical neuroscience Duchet, B Mathematical models of pathological oscillatory activity and effects of deep brain stimulation in movement disorders |
| title | Mathematical models of pathological oscillatory activity and effects of deep brain stimulation in movement disorders |
| title_full | Mathematical models of pathological oscillatory activity and effects of deep brain stimulation in movement disorders |
| title_fullStr | Mathematical models of pathological oscillatory activity and effects of deep brain stimulation in movement disorders |
| title_full_unstemmed | Mathematical models of pathological oscillatory activity and effects of deep brain stimulation in movement disorders |
| title_short | Mathematical models of pathological oscillatory activity and effects of deep brain stimulation in movement disorders |
| title_sort | mathematical models of pathological oscillatory activity and effects of deep brain stimulation in movement disorders |
| topic | Computational neuroscience Neurology Mathematical neuroscience |
| work_keys_str_mv | AT duchetb mathematicalmodelsofpathologicaloscillatoryactivityandeffectsofdeepbrainstimulationinmovementdisorders |