Closed-loop stimulation of a delayed neural fields model of parkinsonian STN-GPe network: a theoretical and computational study

Evoked resonant neural activity long-term dynamics can be reproduced by a computational model with vesicle depletion

Subthalamic deep brain stimulation (DBS) robustly generates high-frequency oscillations known as evoked resonant neural activity (ERNA). Recently the importance of ERNA has been demonstrated through its ability to predict the optimal DBS contact in the subthalamic nucleus in patients with Parkinson's disease. However, the underlying mechanisms of ERNA are not well understood, and previous modelling efforts have not managed to reproduce the wealth of published data describing the dynamics of ERNA. Here, we aim to present a minimal model capable of reproducing the characteristics of the slow ERNA dynamics published to date. We make biophysically-motivated modifications to the Kuramoto model and fit its parameters to the slow dynamics of ERNA obtained from data. Our results demonstrate that it is possible to reproduce the slow dynamics of ERNA (over hundreds of seconds) with a single neuronal population, and, crucially, with vesicle depletion as one of the key mechanisms behind the ERNA frequency decay in our model. We further validate the proposed model against experimental data from Parkinson's disease patients, where it captures the variations in ERNA frequency and amplitude in response to variable stimulation frequency, amplitude, and to stimulation pulse bursting. We provide a series of predictions from the model that could be the subject of future studies for further validation.

Open Hardware Implementation of Real-Time Phase and Amplitude Estimation for Neurophysiologic Signals

Adaptive deep brain stimulation (aDBS) is a promising concept in the field of DBS that consists of delivering electrical stimulation in response to specific events. Dynamic adaptivity arises when stimulation targets dynamically changing states, which often calls for a reliable and fast causal estimation of the phase and amplitude of the signals. Here, we present an open-hardware implementation that exploits the concepts of resonators and Hilbert filters embedded in an open-hardware platform. To emulate real-world scenarios, we built a hardware setup that included a system to replay and process different types of physiological signals and test the accuracy of the instantaneous phase and amplitude estimates. The results show that the system can provide a precise and reliable estimation of the phase even in the challenging scenario of dealing with high-frequency oscillations (~250 Hz) in real-time. The framework might be adopted in neuromodulation studies to quickly test biomarkers in clinical and preclinical settings, supporting the advancement of aDBS.

Efficient suppression of parkinsonian beta oscillations in a closed-loop model of deep brain stimulation with amplitude modulation

Introduction

Parkinson's disease (PD) is a movement disorder characterized by the pathological beta band (15-30 Hz) neural oscillations within the basal ganglia (BG). It is shown that the suppression of abnormal beta oscillations is correlated with the improvement of PD motor symptoms, which is a goal of standard therapies including deep brain stimulation (DBS). To overcome the stimulation-induced side effects and inefficiencies of conventional DBS (cDBS) and to reduce the administered stimulation current, closed-loop adaptive DBS (aDBS) techniques were developed. In this method, the frequency and/or amplitude of stimulation are modulated based on various disease biomarkers.

Methods

Here, by computational modeling of a cortico-BG-thalamic network in normal and PD conditions, we show that closed-loop aDBS of the subthalamic nucleus (STN) with amplitude modulation leads to a more effective suppression of pathological beta oscillations within the parkinsonian BG.

Results

Our results show that beta band neural oscillations are restored to their normal range and the reliability of the response of the thalamic neurons to motor cortex commands is retained due to aDBS with amplitude modulation. Furthermore, notably less stimulation current is administered during aDBS compared with cDBS due to a closed-loop control of stimulation amplitude based on the STN local field potential (LFP) beta activity.

Discussion

Efficient models of closed-loop stimulation may contribute to the clinical development of optimized aDBS techniques designed to reduce potential stimulation-induced side effects of cDBS in PD patients while leading to a better therapeutic outcome.

A bidirectional Hopf bifurcation analysis of Parkinson's oscillation in a simplified basal ganglia model

In this paper, we study the parkinson oscillation mechanism in a computational model by bifurcation analysis and numerical simulation. Oscillatory activities can be induced by abnormal coupling weights and delays. The bidirectional Hopf bifurcation phenomena are found in simulations, which can uniformly explain the oscillation mechanism in this model. The Hopf1 represents the transition between the low firing rate stable state (SS) and oscillation state (OS), the Hopf2 represents the transition between the high firing rate stable state (HSS) and the OS, the mechanisms of them are different. The Hopf1 and Hopf2 bifurcations both show that when the state transfers from the stable region to the oscillation region, oscillatory activities always originate from the beta frequency band, and then gradually evolve into the alpha frequency band, the theta frequency band and delta frequency band in this model. We find that the changing trends of the DF and oscillation amplitude (OSAM) are contrary, oscillation activities in lower frequency band are more stable than that in higher frequency band. The effect of the delay in inhibitory pathways is greater than that of in excitatory pathways, and appropriate delays improve the discharge activation level (DAL) of the system. In all, we infer that oscillations can be induced by the follow factors: 1. Improvement of the DAL of the globus pallidus externa (GPe); 2. Reduce the DAL of the GPe from the HSS or the discharge saturation state; 3. The GPe can also resonate with the subthalamic nucleus (STN).

Closing the loop of DBS using the beta oscillations in cortex

Cortical information has great importance to reflect the deep brain stimulation (DBS) effects for Parkinson's disease patients. Using cortical activities to feedback is an available closed-loop idea for DBS. Previous studies have demonstrated the pathological beta (12-35 Hz) cortical oscillations can be suppressed by appropriate DBS settings. Thus, here we propose to close the loop of DBS based on the beta oscillations in cortex. By modify the cortico-basal ganglia-thalamic neural loop model, more biologically realistic underlying the Parkinsonian phenomenon is approached. Stimulation results show the proposed closed-loop DBS strategy using cortical beta oscillation as feedback information has more profound roles in alleviating the pathological neural abnormality than the traditional open-loop DBS. Additionally, we compare the stimulation effects with subthalamic nucleus feedback strategy. It is shown that using cortical beta information as the feedback signals can further enlarge the control parameter space based on proportional-integral control structure with a lower energy expenditure. This work may pave the way to optimizing the DBS effects in a closed-loop arrangement.

Simulation of Closed-Loop Deep Brain Stimulation Control Schemes for Suppression of Pathological Beta Oscillations in Parkinson's Disease

This study presents a computational model of closed-loop control of deep brain stimulation (DBS) for Parkinson's disease (PD) to investigate clinically viable control schemes for suppressing pathological beta-band activity. Closed-loop DBS for PD has shown promising results in preliminary clinical studies and offers the potential to achieve better control of patient symptoms and side effects with lower power consumption than conventional open-loop DBS. However, extensive testing of algorithms in patients is difficult. The model presented provides a means to explore a range of control algorithms in silico and optimize control parameters before preclinical testing. The model incorporates (i) the extracellular DBS electric field, (ii) antidromic and orthodromic activation of STN afferent fibers, (iii) the LFP detected at non-stimulating contacts on the DBS electrode and (iv) temporal variation of network beta-band activity within the thalamo-cortico-basal ganglia loop. The performance of on-off and dual-threshold controllers for suppressing beta-band activity by modulating the DBS amplitude were first verified, showing levels of beta suppression and reductions in power consumption comparable with previous clinical studies. Proportional (P) and proportional-integral (PI) closed-loop controllers for amplitude and frequency modulation were then investigated. A simple tuning rule was derived for selecting effective PI controller parameters to target long duration beta bursts while respecting clinical constraints that limit the rate of change of stimulation parameters. Of the controllers tested, PI controllers displayed superior performance for regulating network beta-band activity whilst accounting for clinical considerations. Proportional controllers resulted in undesirable rapid fluctuations of the DBS parameters which may exceed clinically tolerable rate limits. Overall, the PI controller for modulating DBS frequency performed best, reducing the mean error by 83% compared to DBS off and the mean power consumed to 25% of that utilized by open-loop DBS. The network model presented captures sufficient physiological detail to act as a surrogate for preclinical testing of closed-loop DBS algorithms using a clinically accessible biomarker, providing a first step for deriving and testing novel, clinically suitable closed-loop DBS controllers.

A neural mass model to predict electrical stimulation evoked responses in human and non-human primate brain

OBJECTIVE

Deep brain stimulation (DBS) is a valuable tool for ameliorating drug resistant pathologies such as movement disorders and epilepsy. DBS is also being considered for complex neuro-psychiatric disorders, which are characterized by high variability in symptoms and slow responses that hinder DBS setting optimization. The objective of this work was to develop an in silico platform to examine the effects of electrical stimulation in regions neighboring a stimulated brain region.

APPROACH

We used the Jansen-Rit neural mass model of single and coupled nodes to simulate the response to a train of electrical current pulses at different frequencies (10-160 Hz) of the local field potential recorded in the amygdala and cortical structures in human subjects and a non-human primate.

RESULTS

We found that using a single node model, the evoked responses could be accurately modeled following a narrow range of stimulation frequencies. Including a second coupled node increased the range of stimulation frequencies whose evoked responses could be efficiently modeled. Furthermore, in a chronic recording from a non-human primate, features of the in vivo evoked response remained consistent for several weeks, suggesting that model re-parameterization for chronic stimulation protocols would be infrequent.

SIGNIFICANCE

Using a model of neural population activity, we reproduced the evoked response to cortical and subcortical stimulation in human and non-human primate. This modeling framework provides an environment to explore, safely and rapidly, a wide range of stimulation settings not possible in human brain stimulation studies. The model can be trained on a limited dataset of stimulation responses to develop an optimal stimulation strategy for an individual patient.

Topographic anatomy of the subthalamic nucleus localized by high-resolution human brain atlas superimposing digital images of cross-sectioned surfaces and histological images of microscopic sections from frozen cadaveric brains

Despite the recent advent of neuro-radiographic techniques, creating a 'perfect' human brain atlas providing precise and consistent images with minimal distortion is practically difficult. In this study, we created a new human brain atlas from cadaveric brains with serial sections of 50 μm thickness covering the entire basal ganglia. Human cerebral hemispheres were obtained from 10 donated cadavers and fixed in 10% formalin solution, cut in a block measuring 50 mm × 30 mm × 50 mm around the midpoint of the anterior and posterior commissures and frozen at -40 °C. Each block was cut into 50-μm-thick sections on the freezing microtome and the cross-sectioned surface was photographed. Simultaneously, every 10th slice from one sagittal hemisphere was sampled and stained using the Kluver-Barrera method. Prepared slides were photographed under light microscopy, and data from digital images of the cross-sectioned surface (DICSS) and digital images from microscopic sections (DIMS) were processed. Gray areas on DICSS largely represented areas of dense cellularity, and around subthalamic nucleus (STN), the zona incerta and field of Forel were clearly distinguishable on the anterosuperior side, as was the substantia nigra on the caudal side. DICSS successfully delineated the anatomical structure identical to the STN and surrounding contiguous nuclei. This new brain atlas will allow elucidation of anatomy that cannot be clearly disclosed from modern radiographic imaging or is very difficult to analyze with spatially inconsistent histological sections, and will contribute to further progress in anatomical studies of the human basal ganglia.

Towards miniaturized closed-loop optogenetic stimulation devices

OBJECTIVE

Electrical brain stimulation provides therapeutic benefits for patients with drug-resistant neurological disorders. It, however, has restricted access to cell-type selectivity which limits its treatment effectiveness. Optogenetics, in contrast, enables precise targeting of a specific cell type which can address the issue with electrical brain stimulation. It, nonetheless, disregards real-time brain responses in delivering optimized stimulation to target cells. Closed-loop optogenetics, on the other hand, senses the difference between normal and abnormal states of the brain, and modulates stimulation parameters to achieve the desired stimulation outcome. Current review articles on closed-loop optogenetics have focused on its theoretical aspects and potential benefits. A review of the recent progress in miniaturized closed-loop optogenetic stimulation devices is thus needed.

APPROACH

This paper presents a comprehensive study on the existing miniaturized closed-loop optogenetic stimulation devices and their internal components.

MAIN RESULTS

Hardware components of closed-loop optogenetic stimulation devices including electrode, light-guiding mechanism, optical source, neural recorder, and optical stimulator are discussed. Next, software modules of closed-loop optogenetic stimulation devices including feature extraction, classification, control, and stimulation parameter modulation are described. Then, the existing devices are categorized into open-loop and closed-loop groups, and the combined operation of their neural recorder, optical stimulator, and control approach is discussed. Finally, the challenges in the design and implementation of closed-loop optogenetic stimulation devices are presented, suggestions on how to tackle these challenges are given, and future directions for closed-loop optogenetics are stated.

SIGNIFICANCE

A generic architecture for closed-loop optogenetic stimulation devices involving both hardware and software perspectives is devised. A comprehensive investigation into the most current miniaturized and tetherless closed-loop optogenetic stimulation devices is given. A detailed comparison of the closed-loop optogenetic stimulation devices is presented.

Computational Stimulation of the Basal Ganglia Neurons with Cost Effective Delayed Gaussian Waveforms

Deep brain stimulation (DBS) has compelling results in the desynchronization of the basal ganglia neuronal activities and thus, is used in treating the motor symptoms of Parkinson's disease (PD). Accurate definition of DBS waveform parameters could avert tissue or electrode damage, increase the neuronal activity and reduce energy cost which will prolong the battery life, hence avoiding device replacement surgeries. This study considers the use of a charge balanced Gaussian waveform pattern as a method to disrupt the firing patterns of neuronal cell activity. A computational model was created to simulate ganglia cells and their interactions with thalamic neurons. From the model, we investigated the effects of modified DBS pulse shapes and proposed a delay period between the cathodic and anodic parts of the charge balanced Gaussian waveform to desynchronize the firing patterns of the GPe and GPi cells. The results of the proposed Gaussian waveform with delay outperformed that of rectangular DBS waveforms used in in-vivo experiments. The Gaussian Delay Gaussian (GDG) waveforms achieved lower number of misses in eliciting action potential while having a lower amplitude and shorter length of delay compared to numerous different pulse shapes. The amount of energy consumed in the basal ganglia network due to GDG waveforms was dropped by 22% in comparison with charge balanced Gaussian waveforms without any delay between the cathodic and anodic parts and was also 60% lower than a rectangular charged balanced pulse with a delay between the cathodic and anodic parts of the waveform. Furthermore, by defining a Synchronization Level metric, we observed that the GDG waveform was able to reduce the synchronization of GPi neurons more effectively than any other waveform. The promising results of GDG waveforms in terms of eliciting action potential, desynchronization of the basal ganglia neurons and reduction of energy consumption can potentially enhance the performance of DBS devices.

Portable closed-loop optogenetic stimulation device

This paper presents a closed-loop optogenetic stimulation device to achieve online modulation of neurons. The device is designed to be mountable on small rodents in pre-clinical settings. Considering the size of rodents and the need for portability, a single-piece self-contained device is developed which allows real-time photostimulation based on detected neuronal states. It consists of three components: a neural recorder, a control algorithm, and an optogenetic stimulator. The neural recorder which is realized by analogue circuitry measures the neural signal. The on-off control algorithm analyses the neural signal and controls the stimulation of the target neurons. The optogenetic stimulator performs sampling and digitization of the detected neural signal, runs the control algorithm, and manages the operation of the light source. The configurable neural recorder is capable of 64 dB amplification in the frequency range of 300 Hz to 6 KHz. The outcome of bench testing of the device is reported. The device is portable and headmountable which makes it suitable for use with small rodents in pre-clinical trials.

Bio-Inspired Controller on an FPGA Applied to Closed-Loop Diaphragmatic Stimulation

Cervical spinal cord injury can disrupt connections between the brain respiratory network and the respiratory muscles which can lead to partial or complete loss of ventilatory control and require ventilatory assistance. Unlike current open-loop technology, a closed-loop diaphragmatic pacing system could overcome the drawbacks of manual titration as well as respond to changing ventilation requirements. We present an original bio-inspired assistive technology for real-time ventilation assistance, implemented in a digital configurable Field Programmable Gate Array (FPGA). The bio-inspired controller, which is a spiking neural network (SNN) inspired by the medullary respiratory network, is as robust as a classic controller while having a flexible, low-power and low-cost hardware design. The system was simulated in MATLAB with FPGA-specific constraints and tested with a computational model of rat breathing; the model reproduced experimentally collected respiratory data in eupneic animals. The open-loop version of the bio-inspired controller was implemented on the FPGA. Electrical test bench characterizations confirmed the system functionality. Open and closed-loop paradigm simulations were simulated to test the FPGA system real-time behavior using the rat computational model. The closed-loop system monitors breathing and changes in respiratory demands to drive diaphragmatic stimulation. The simulated results inform future acute animal experiments and constitute the first step toward the development of a neuromorphic, adaptive, compact, low-power, implantable device. The bio-inspired hardware design optimizes the FPGA resource and time costs while harnessing the computational power of spike-based neuromorphic hardware. Its real-time feature makes it suitable for in vivo applications.