Classification of pallidal oscillations with increasing parkinsonian severity

Machine Learning Identifies a Rat Model of Parkinson's Disease via Sleep-Wake Electroencephalogram

Alpha-synuclein induced degeneration of the midbrain substantia nigra pars compact (SNc) dopaminergic neurons causes Parkinson's disease (PD). Rodent studies demonstrate that nigrostriatal dopamine stimulates pallidal neurons which, via the topographical pallidocortical pathway, regulate cortical activity and functions. We hypothesize that nigrostriatal dopamine acting at the basal ganglia regulates cortical activity in sleep and wake state, and its depletion systemically alters electroencephalogram (EEG) across frequencies during sleep-wake state. Compared to control rats, 6-hydroxydopamine induced selective SNc lesions increased overall EEG power (positive synchronization) across 0.5-60 Hz during wake, NREM (non-rapid eye movement) sleep, and REM sleep. Application of machine learning (ML) to seven EEG features computed at a single or combined spectral bands during sleep-wake differentiated SNc lesions from controls at high accuracy. ML algorithms construct a model based on empirical data to make predictions on subsequent data. The accuracy of the predictive results indicate that nigrostriatal dopamine depletion increases global EEG spectral synchronization in wake, NREM sleep, and REM sleep. The EEG changes can be exploited by ML to identify SNc lesions at a high accuracy.

Parkinsonian daytime sleep-wake classification using deep brain stimulation lead recordings

Excessive daytime sleepiness is a recognized non-motor symptom that adversely impacts the quality of life of people with Parkinson's disease (PD), yet effective treatment options remain limited. Deep brain stimulation (DBS) of the subthalamic nucleus (STN) is an effective treatment for PD motor signs. Reliable daytime sleep-wake classification using local field potentials (LFPs) recorded from DBS leads implanted in STN can inform the development of closed-loop DBS approaches for prompt detection and disruption of sleep-related neural oscillations. We performed STN DBS lead recordings in three nonhuman primates rendered parkinsonian by administrating neurotoxin 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP). Reference sleep-wake states were determined on a second-by-second basis by video monitoring of eyes (eyes-open, wake and eyes-closed, sleep). The spectral power in delta (1-4 Hz), theta (4-8 Hz), low-beta (8-20 Hz), high-beta (20-35 Hz), gamma (35-90 Hz), and high-frequency (200-400 Hz) bands were extracted from each wake and sleep epochs for training (70% data) and testing (30% data) a support vector machines classifier for each subject independently. The spectral features yielded reasonable daytime sleep-wake classification (sensitivity: 90.68 ± 1.28; specificity: 88.16 ± 1.08; accuracy: 89.42 ± 0.68; positive predictive value; 88.70 ± 0.89, n = 3). Our findings support the plausibility of monitoring daytime sleep-wake states using DBS lead recordings. These results could have future clinical implications in informing the development of closed-loop DBS approaches for automatic detection and disruption of sleep-related neural oscillations in people with PD to promote wakefulness.

Linking α-synuclein-induced synaptopathy and neural network dysfunction in early Parkinson's disease

The prodromal phase of Parkinson's disease is characterized by aggregation of the misfolded pathogenic protein α-synuclein in select neural centres, co-occurring with non-motor symptoms including sensory and cognitive loss, and emotional disturbances. It is unclear whether neuronal loss is significant during the prodrome. Underlying these symptoms are synaptic impairments and aberrant neural network activity. However, the relationships between synaptic defects and network-level perturbations are not established. In experimental models, pathological α-synuclein not only impacts neurotransmission at the synaptic level, but also leads to changes in brain network-level oscillatory dynamics-both of which likely contribute to non-motor deficits observed in Parkinson's disease. Here we draw upon research from both human subjects and experimental models to propose a 'synapse to network prodrome cascade' wherein before overt cell death, pathological α-synuclein induces synaptic loss and contributes to aberrant network activity, which then gives rise to prodromal symptomology. As the disease progresses, abnormal patterns of neural activity ultimately lead to neuronal loss and clinical progression of disease. Finally, we outline goals and research needed to unravel the basis of functional impairments in Parkinson's disease and other α-synucleinopathies.

Controlling pallidal oscillations in real-time in Parkinson's disease using evoked interference deep brain stimulation (eiDBS): Proof of concept in the human

Approaches to control basal ganglia neural activity in real-time are needed to clarify the causal role of 13-35 Hz ("beta band") oscillatory dynamics in the manifestation of Parkinson's disease (PD) motor signs. Here, we show that resonant beta oscillations evoked by electrical pulses with precise amplitude and timing can be used to predictably suppress or amplify spontaneous beta band activity in the internal segment of the globus pallidus (GPi) in the human. Using this approach, referred to as closed-loop evoked interference deep brain stimulation (eiDBS), we could suppress or amplify frequency-specific (16-22 Hz) neural activity in a PD patient. Our results highlight the utility of eiDBS to characterize the role of oscillatory dynamics in PD and other brain conditions, and to develop personalized neuromodulation systems.

Modulation of dopamine tone induces frequency shifts in cortico-basal ganglia beta oscillations

Βeta oscillatory activity (human: 13-35 Hz; primate: 8-24 Hz) is pervasive within the cortex and basal ganglia. Studies in Parkinson's disease patients and animal models suggest that beta-power increases with dopamine depletion. However, the exact relationship between oscillatory power, frequency and dopamine tone remains unclear. We recorded neural activity in the cortex and basal ganglia of healthy non-human primates while acutely and chronically up- and down-modulating dopamine levels. We assessed changes in beta oscillations in patients with Parkinson's following acute and chronic changes in dopamine tone. Here we show beta oscillation frequency is strongly coupled with dopamine tone in both monkeys and humans. Power, coherence between single-units and local field potentials (LFP), spike-LFP phase-locking, and phase-amplitude coupling are not systematically regulated by dopamine levels. These results demonstrate that beta frequency is a key property of pathological oscillations in cortical and basal ganglia networks.

Revisiting the "Paradox of Stereotaxic Surgery": Insights Into Basal Ganglia-Thalamic Interactions

Basal ganglia dysfunction is implicated in movement disorders including Parkinson Disease, dystonia, and choreiform disorders. Contradicting standard "rate models" of basal ganglia-thalamic interactions, internal pallidotomy improves both hypo- and hyper-kinetic movement disorders. This "paradox of stereotaxic surgery" was recognized shortly after rate models were developed, and is underscored by the outcomes of deep brain stimulation (DBS) for movement disorders. Despite strong evidence that DBS activates local axons, the clinical effects of lesions and DBS are nearly identical. These observations argue against standard models in which GABAergic basal ganglia output gates thalamic activity, and raise the question of how lesions and stimulation can have similar effects. These paradoxes may be resolved by considering thalamocortical loops as primary drivers of motor output. Rather than suppressing or releasing cortex via motor thalamus, the basal ganglia may modulate the timing of thalamic perturbations to cortical activity. Motor cortex exhibits rotational dynamics during movement, allowing the same thalamocortical perturbation to affect motor output differently depending on its timing with respect to the rotational cycle. We review classic and recent studies of basal ganglia, thalamic, and cortical physiology to propose a revised model of basal ganglia-thalamocortical function with implications for basic physiology and neuromodulation.

Longitudinal analysis of local field potentials recorded from directional deep brain stimulation lead implants in the subthalamic nucleus

Objective.The electrode-tissue interface surrounding a deep brain stimulation (DBS) lead is known to be highly dynamic following implantation, which may have implications on the interpretation of intraoperatively recorded local field potentials (LFPs). We characterized beta-band LFP dynamics following implantation of a directional DBS lead in the sensorimotor subthalamic nucleus (STN), which is a primary target for treating Parkinson's disease.Approach.Directional STN-DBS leads were implanted in four healthy, non-human primates. LFPs were recorded over two weeks and again 1-4 months after implantation. Impedance was measured for two weeks post-implant without stimulation to compare the reactive tissue response to changes in LFP oscillations. Beta-band (12-30 Hz) peak power was calculated from the LFP power spectra using both common average referencing (CAR) and intra-row bipolar referencing (IRBR).Results.Resting-state LFPs in two of four subjects revealed a steady increase of beta power over the initial two weeks post-implant whereas the other two subjects showed variable changes over time. Beta power variance across days was significantly larger in the first two weeks compared to 1-4 months post-implant in all three long-term subjects. Further, spatial maps of beta power several hours after implantation did not correlate with those measured two weeks or 1-4 months post-implant. CAR and IRBR beta power correlated across short- and long-term time points. However, depending on the time period, subjects showed a significant bias towards larger beta power using one referencing scheme over the other. Lastly, electrode-tissue impedance increased over the two weeks post-implant but showed no significant correlation to beta power.Significance.These results suggest that beta power in the STN may undergo significant changes following DBS lead implantation. DBS lead diameter and electrode recording configurations can affect the post-implant interpretation of oscillatory features. Such insights will be important for extrapolating results from intraoperative and externalized LFP recordings.

A Pilot Study on Data-Driven Adaptive Deep Brain Stimulation in Chronically Implanted Essential Tremor Patients

Deep brain stimulation (DBS) is an established therapy for Parkinson's disease (PD) and essential-tremor (ET). In adaptive DBS (aDBS) systems, online tuning of stimulation parameters as a function of neural signals may improve treatment efficacy and reduce side-effects. State-of-the-art aDBS systems use symptom surrogates derived from neural signals-so-called neural markers (NMs)-defined on the patient-group level, and control strategies assuming stationarity of symptoms and NMs. We aim at improving these aDBS systems with (1) a data-driven approach for identifying patient- and session-specific NMs and (2) a control strategy coping with short-term non-stationary dynamics. The two building blocks are implemented as follows: (1) The data-driven NMs are based on a machine learning model estimating tremor intensity from electrocorticographic signals. (2) The control strategy accounts for local variability of tremor statistics. Our study with three chronically implanted ET patients amounted to five online sessions. Tremor quantified from accelerometer data shows that symptom suppression is at least equivalent to that of a continuous DBS strategy in 3 out-of 4 online tests, while considerably reducing net stimulation (at least 24%). In the remaining online test, symptom suppression was not significantly different from either the continuous strategy or the no treatment condition. We introduce a novel aDBS system for ET. It is the first aDBS system based on (1) a machine learning model to identify session-specific NMs, and (2) a control strategy coping with short-term non-stationary dynamics. We show the suitability of our aDBS approach for ET, which opens the door to its further study in a larger patient population.

Machine Learning's Application in Deep Brain Stimulation for Parkinson's Disease: A Review

Deep brain stimulation (DBS) is a surgical treatment for advanced Parkinson's disease (PD) that has undergone technological evolution that parallels an expansion in clinical phenotyping, neurophysiology, and neuroimaging of the disease state. Machine learning (ML) has been successfully used in a wide range of healthcare problems, including DBS. As computational power increases and more data become available, the application of ML in DBS is expected to grow. We review the literature of ML in DBS and discuss future opportunities for such applications. Specifically, we perform a comprehensive review of the literature from PubMed, the Institute for Scientific Information's Web of Science, Cochrane Database of Systematic Reviews, and Institute of Electrical and Electronics Engineers' (IEEE) Xplore Digital Library for ML applications in DBS. These studies are broadly placed in the following categories: (1) DBS candidate selection; (2) programming optimization; (3) surgical targeting; and (4) insights into DBS mechanisms. For each category, we provide and contextualize the current body of research and discuss potential future directions for the application of ML in DBS.

Changing views of the pathophysiology of Parkinsonism

Studies of the pathophysiology of parkinsonism (specifically akinesia and bradykinesia) have a long history and primarily model the consequences of dopamine loss in the basal ganglia on the function of the basal ganglia/thalamocortical circuit(s). Changes of firing rates of individual nodes within these circuits were originally considered central to parkinsonism. However, this view has now given way to the belief that changes in firing patterns within the basal ganglia and related nuclei are more important, including the emergence of burst discharges, greater synchrony of firing between neighboring neurons, oscillatory activity patterns, and the excessive coupling of oscillatory activities at different frequencies. Primarily focusing on studies obtained in nonhuman primates and human patients with Parkinson's disease, this review summarizes the current state of this field and highlights several emerging areas of research, including studies of the impact of the heterogeneity of external pallidal neurons on parkinsonism, the importance of extrastriatal dopamine loss, parkinsonism-associated synaptic and morphologic plasticity, and the potential role(s) of the cerebellum and brainstem in the motor dysfunction of Parkinson's disease. © 2019 International Parkinson and Movement Disorder Society.

State transitions in the substantia nigra reticulata predict the onset of motor deficits in models of progressive dopamine depletion in mice

Parkinson's disease (PD) is a progressive neurodegenerative disorder whose cardinal motor symptoms are attributed to dysfunction of basal ganglia circuits under conditions of low dopamine. Despite well-established physiological criteria to define basal ganglia dysfunction, correlations between individual parameters and motor symptoms are often weak, challenging their predictive validity and causal contributions to behavior. One limitation is that basal ganglia pathophysiology is studied only at end-stages of depletion, leaving an impoverished understanding of when deficits emerge and how they evolve over the course of depletion. In this study, we use toxin- and neurodegeneration-induced mouse models of dopamine depletion to establish the physiological trajectory by which the substantia nigra reticulata (SNr) transitions from the healthy to the diseased state. We find that physiological progression in the SNr proceeds in discrete state transitions that are highly stereotyped across models and correlate well with the prodromal and symptomatic stages of behavior.

High-resolution local field potentials measured with deep brain stimulation arrays

OBJECTIVE

Local field potential (LFP) recordings along a deep brain stimulation (DBS) lead can provide useful feedback for titrating DBS therapy. However, conventional DBS leads with four cylindrical macroelectrodes likely undersample the spatial distribution of sinks and sources in a given brain region. In this study, we investigated the spectral power and spatial feature sizes of LFP activity in non-human primate subthalamic nucleus and globus pallidus using chronically implanted 32-channel directional DBS arrays.

APPROACH

Subthalamic nucleus and globus pallidus LFP signals were recorded from directional DBS arrays in the resting state and during a reach-and-retrieval task in two non-human primates in naïve and parkinsonian conditions. LFP recordings were compared amongst bipolar pairs of electrodes using individual and grouped electrode configurations, with the latter mimicking the cylindrical macroelectrode configurations used in current clinical LFP recordings.

MAIN RESULTS

Recordings from these DBS arrays showed that (1) beta oscillations have spatial 'fingerprints' in the subthalamic nucleus and globus pallidus, and (2) that these oscillations were muted when grouping electrode contacts together to create cylindrical macroelectrodes similar in relative dimension to those used clinically. Further, these maps depended on parkinsonian condition and whether the subject was resting or performing a motor task.

SIGNIFICANCE

Development of future closed-loop DBS therapies that rely on LFP feedback will benefit from implanting DBS arrays with electrode sizes and spacings that are more consistent with the dimensions of oscillatory sinks and sources within the brain.

Algorithmic design of a noise-resistant and efficient closed-loop deep brain stimulation system: A computational approach

Advances in the field of closed-loop neuromodulation call for analysis and modeling approaches capable of confronting challenges related to the complex neuronal response to stimulation and the presence of strong internal and measurement noise in neural recordings. Here we elaborate on the algorithmic aspects of a noise-resistant closed-loop subthalamic nucleus deep brain stimulation system for advanced Parkinson’s disease and treatment-refractory obsessive-compulsive disorder, ensuring remarkable performance in terms of both efficiency and selectivity of stimulation, as well as in terms of computational speed. First, we propose an efficient method drawn from dynamical systems theory, for the reliable assessment of significant nonlinear coupling between beta and high-frequency subthalamic neuronal activity, as a biomarker for feedback control. Further, we present a model-based strategy through which optimal parameters of stimulation for minimum energy desynchronizing control of neuronal activity are being identified. The strategy integrates stochastic modeling and derivative-free optimization of neural dynamics based on quadratic modeling. On the basis of numerical simulations, we demonstrate the potential of the presented modeling approach to identify, at a relatively low computational cost, stimulation settings potentially associated with a significantly higher degree of efficiency and selectivity compared with stimulation settings determined post-operatively. Our data reinforce the hypothesis that model-based control strategies are crucial for the design of novel stimulation protocols at the backstage of clinical applications.

Particle swarm optimization for programming deep brain stimulation arrays

OBJECTIVE

Deep brain stimulation (DBS) therapy relies on both precise neurosurgical targeting and systematic optimization of stimulation settings to achieve beneficial clinical outcomes. One recent advance to improve targeting is the development of DBS arrays (DBSAs) with electrodes segmented both along and around the DBS lead. However, increasing the number of independent electrodes creates the logistical challenge of optimizing stimulation parameters efficiently.

APPROACH

Solving such complex problems with multiple solutions and objectives is well known to occur in biology, in which complex collective behaviors emerge out of swarms of individual organisms engaged in learning through social interactions. Here, we developed a particle swarm optimization (PSO) algorithm to program DBSAs using a swarm of individual particles representing electrode configurations and stimulation amplitudes. Using a finite element model of motor thalamic DBS, we demonstrate how the PSO algorithm can efficiently optimize a multi-objective function that maximizes predictions of axonal activation in regions of interest (ROI, cerebellar-receiving area of motor thalamus), minimizes predictions of axonal activation in regions of avoidance (ROA, somatosensory thalamus), and minimizes power consumption.

MAIN RESULTS

The algorithm solved the multi-objective problem by producing a Pareto front. ROI and ROA activation predictions were consistent across swarms (<1% median discrepancy in axon activation). The algorithm was able to accommodate for (1) lead displacement (1 mm) with relatively small ROI (⩽9.2%) and ROA (⩽1%) activation changes, irrespective of shift direction; (2) reduction in maximum per-electrode current (by 50% and 80%) with ROI activation decreasing by 5.6% and 16%, respectively; and (3) disabling electrodes (n  =  3 and 12) with ROI activation reduction by 1.8% and 14%, respectively. Additionally, comparison between PSO predictions and multi-compartment axon model simulations showed discrepancies of  <1% between approaches.

SIGNIFICANCE

The PSO algorithm provides a computationally efficient way to program DBS systems especially those with higher electrode counts.

Physiological changes in the pallidum in a progressive model of Parkinson's disease: Are oscillations enough?

Neurophysiological changes in the basal ganglia thalamo-cortical circuit associated with the development of parkinsonian motor signs remain poorly understood. Theoretical models have ranged from those emphasizing changes in mean discharge rate to increased oscillatory activity within the beta range. The present study characterized neuronal activity within and across the internal and external segments of the globus pallidus as a function of motor severity using a staged, progressively severe 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine model of Parkinsonism in three rhesus monkeys. An increase in coherence between neuronal pairs across the external and internal globus pallidus was present in multiple frequency bands in the parkinsonian state; both the peak frequency of oscillatory coherence and the variability were reduced in the parkinsonian state. The incidence of 8-20Hz oscillatory activity in the internal globus pallidus increased with the progression of the disease when pooling the data across the three animals; however it did not correlate with motor severity when assessed individually and increased progressively in only one of three animals. No systematic relationship between mean discharge rates or the incidence or structure of bursting activity and motor severity was observed. These data suggest that exaggerated coupling across pallidal segments contribute to the development of the parkinsonian state by inducing an exaggerated level of synchrony and loss of focusing within the basal ganglia. These data further point to the lack of a defined relationship between rate changes, the mere presence of oscillatory activity in the beta range and bursting activity in the basal ganglia to the motor signs of Parkinson's disease.

Phase-amplitude coupling, an indication of bursting in parkinsonism, is masked by periodic pulses

Interactions between neural oscillations in the brain have been observed in many structures including the hippocampus, amygdala, motor cortex, and basal ganglia. In this study, one popular approach for quantifying oscillation interactions was considered: phase-amplitude coupling. The goals of the study were to use simulations to examine potential causes of elevated phase-amplitude coupling in parkinsonism, to compare simulated parkinsonian signals with recorded local field potentials from animal models of parkinsonism, to investigate possible relationships between increased bursting in parkinsonian single cells and elevated phase-amplitude coupling, and to uncover potential noise and artifact effects. First, a cell model that integrates incremental input currents and fires at realistic voltage thresholds was modified to allow control of stochastic parameters related to firing and burst rates. Next, the input currents and distribution of integration times were set to reproduce firing patterns consistent with those from parkinsonian subthalamic nucleus cells. Then, local field potentials were synthesized from the output of multiple simulated cells with varying degrees of synchronization and compared with subthalamic nucleus recordings from animal models of parkinsonism. The results showed that phase-amplitude coupling can provide important information about underlying neural activity. In particular, signals synthesized from synchronized bursting neurons showed increased oscillatory interactions similar to those observed in parkinsonian animals. Additionally, changes in bursting parameters such as the intraburst rate, the mean interburst period, and the amount of synchronization between neurons influenced the phase-amplitude coupling in predictable ways. Finally, simulation results revealed that small periodic signals can have a surprisingly large masking effect on phase-amplitude coupling.