Clare M. Davidson, Annraoi M. de Paor, Hayriye Cagnan and Madeleine M. Lowery, University College Dublin, Ireland
Parkinson’s disease is a progressive, neurodegenerative disorder which primarily affects the motor system in humans. Recordings from the basal ganglia area of the brain reveal increased synchronous and oscillatory neural activity in Parkinson’s disease when compared to a healthy brain. Correlation has been established between some of this pathological activity and the motor symptoms of Parkinson’s disease. Deep brain stimulation (DBS) with chronically implanted electrodes is a successful way to control the symptoms of Parkinson’s disease for many people with the condition. It has also been shown to suppress the increased oscillatory neural activity observed in the Parkinsonian basal ganglia at particular frequencies. However, exactly why and how DBS works remains largely unknown. In addition, the optimal stimulation settings vary from person to person. Identifying the stimulation settings that work best for an individual can be a time-consuming and complex process. To better understand the interaction of neural oscillations and their suppression with high frequency DBS, we have developed a mathematical model representing the pathological oscillations that are observed in the Parkinsonian basal ganglia. When we apply a stimulating waveform (similar to DBS) to our model, the amplitude of this oscillatory activity decreases in a frequency and amplitude dependent manner. This phenomenon is also observed in local field potential recordings when DBS is applied to the basal ganglia. Our model can be tuned to match these local field potentials recorded in patients. We show how the stimulation settings required to achieve a specified level of suppression of neural activity can be predicted mathematically. The developed mathematical model provides a way to explore DBS settings in silico.
Keywords: Basal ganglia, mean field model, Parkinson’s disease, pathological oscillations, control theoryREAD FULL ARTICLE ON IEEE XPLORE