Search

Objectives To evaluate the correlation between the pallidal local field potentials (LFPs) activity and the cortical oscillations (at rest and during several motor tasks) in two freely moving patients with generalized dystonia and pallidal deep brain stimulation (DBS). Materials and methods Two women with isolated generalized dystonia were selected for bilateral globus pallidus internus (GPi) DBS. After the electrodes' implantation, cortical activity was recorded by a portable electroencephalography (EEG) system simultaneously with GPi LFPs activity, during several motor tasks, gait, and rest condition. Recordings were not performed during stimulation. EEG and LFPs signals relative to each specific movement were coupled together and grouped in neck/upper limbs movements and gait. Power spectral density (PSD), EEG-LFP coherence (through envelope of imaginary coherence operator), and 1/f exponent of LFP-PSD background were calculated. Results In both patients, the pallidal LFPs PSD at rest was characterized by prominent 4-12 Hz activity. Voluntary movements increased activity in the theta (θ) band (4-7 Hz) compared to rest, in both LFPs and EEG signals. Gait induced a drastic raise of θ activity in both patients' pallidal activity, less marked for the EEG signal. A coherence peak within the 8-13 Hz range was found between pallidal LFPs and EEG recorded at rest. Conclusions Neck/upper limbs voluntary movements and gait suppressed the GPi-LFPs-cortical-EEG coherence and differently impacted both EEG and LFPs low frequency activity. These findings suggest a selective modulation of the cortico-basal ganglia network activity in dystonia.

The process of recording local fields potentials can be influenced by many internal and external sources of electrical noise. To successfully use these recordings, noise must be removed, for which an automatic detection tool is needed to speed up the reviewing process. This research aims to develop an interpretable model, and among the many machine learning models, decision trees stand out due to the innate ability to allow insight into the classification criteria. As they require extracted features instead of the raw signal to achieve good performance, the adaptation of features was proposed originally intended for electroencephalography classification for the detection of artefacts in local fields potentials. Afterwards, they are grouped, and three different filtering feature selection algorithms are applied to obtain the most relevant ones and compare their choices. Classification accuracy of 88.1% is obtained with a single feature. Hence, it is concluded that interpretable models can be obtained with a performance similar to that of deep learning approaches.

Neuronal signals generally represent activation of the neuronal networks and give insights into brain functionalities. They are considered as fingerprints of actions and their processing across different structures of the brain. These recordings generate a large volume of data that are susceptible to noise and artifacts. Therefore, the review of these data to ensure high quality by automatically detecting and removing the artifacts is imperative. Toward this aim, this work proposes a custom-developed automatic artifact removal toolbox named, SANTIA (SigMate Advanced: a Novel Tool for Identification of Artifacts in Neuronal Signals). Developed in Matlab, SANTIA is an open-source toolbox that applies neural network-based machine learning techniques to label and train models to detect artifacts from the invasive neuronal signals known as local field potentials.

Local fields potentials (LFP) can be contaminated by different internal and external sources of noise during their recordings. In cases where artefacts are present in them, automatic detection tools are needed to speed up the high accuracy detection process, followed by their removal to successfully use these recordings. This process is facilitated by a pool of supervised machine learning based tools which require labelled data for training. These algorithms have the capacity to distinguish between normal brain patterns and artefacts from an individual or a group, which is more flexible than template matching and subtraction. In addition, their portability has seen developments from both the software and hardware perspective. For many tools, LFP signal power is used as a gauge to measure and eventually label the artefacts portion of the LFP. This work explores how signal power affects the detection and classification accuracy of artefacts. Results show that a higher threshold value impacts positively the accuracy, due to less false positives in the data, without compromising the specificity in balanced datasets.

Acquired brain injuries, such as stroke, are a major cause of long-term disability worldwide. When an injury occurs within primary motor cortex (M1) there is both a loss of descending information to the spinal cord and a disruption in the communication within and between spared regions, which can lead to hemiparesis and other motor dysfunctions. Recently, activity-dependent stimulation (ADS), a closed loop intracortical neurostimulation technique, was shown to significantly improve behavioural recovery in rodent models following a unilateral traumatic brain injury in the primary motor cortex. While behavioural benefits of this approach have been described, the neurophysiological changes have not yet been fully characterized. In this study, we investigated how neural activity in the form of local field potential (LFP) was impacted by an ischemic lesion and how it changed following prolonged ADS sessions. Intracortical microelectrodes arrays were used to record and stimulate from the rostral forelimb area (RFA) and the primary somatosensory cortex (S1) of 12 rats (Long Evans). An ischemic injury induced with Endothelin-1 microinjections in the caudal forelimb area (CFA) was used as an injury model. We found that the ET-1 caused lesion led to widespread drop of LFP power across all the frequency bands in both the monitored areas. The lesion did not change the functional connectivity within the brain areas, while it increased the synchronization between them. Following stimulation, LFP power generally increased in both areas. Stimulation also led to a decrement in connectivity between S1 and RFA, while causing the within S1 connectivity to increase.

Summary Recent advances in bioelectronics and neural engineering allowed the development of brain machine interfaces and neuroprostheses, capable of facilitating or recovering functionality in people with neurological disability. To realize energy-efficient and real-time capable devices, neuromorphic computing systems are envisaged as the core of next-generation systems for brain repair. We demonstrate here a real-time hardware neuromorphic prosthesis to restore bidirectional interactions between two neuronal populations, even when one is damaged or missing. We used in vitro modular cell cultures to mimic the mutual interaction between neuronal assemblies and created a focal lesion to functionally disconnect the two populations. Then, we employed our neuromorphic prosthesis for bidirectional bridging to artificially reconnect two disconnected neuronal modules and for hybrid bidirectional bridging to replace the activity of one module with a real-time hardware neuromorphic Spiking Neural Network. Our neuroprosthetic system opens avenues for the exploitation of neuromorphic-based devices in bioelectrical therapeutics for health care., Graphical Abstract, Highlights • We developed a novel real-time neuromorphic system acting as a neuroprosthesis • The neuroprosthesis can bidirectionally reconnect two disconnected networks • The neuroprosthesis implements a real-time spiking neural network • The spiking network can successfully replace the activity of a neural population, Neuroscience; Systems Neuroscience; Computer Science; Evolvable Hardware; Electronic Materials

As our understanding of volitional motor function increases, it is clear that complex movements are the result of the interactions of multiple cortical regions rather than just the output properties of primary motor cortex. However, our understanding of the interactions among these regions is limited. In this study, we used the activity-dependent stimulation (ADS) technique to determine the short/long-term effects on network activity and neuroplasticity of intracortical connections. ADS uses the intrinsic neural activity of one region to trigger stimulations in a separate region of the brain and can manipulate neuronal connectivity in vivo. Our aim was to compare single-unit neuronal activity within premotor cortex (rostral forelimb area, [RFA] in rats) in response to ADS (triggered from RFA) and randomly-generated stimulation in the somatosensory area (S1) within single sessions and across 21 consecutive days of stimulation. We examined firing rate and correlation between spikes and stimuli in chronically-implanted healthy ambulatory rats during spontaneous and evoked activity. At the end of the treatment, we evaluated changes of synaptophysin expression. Our results demonstrated the ability of ADS to modulate RFA firing properties and to promote synaptogenesis in S1, strengthening the idea that this Hebbian-inspired protocol can be used to modulate cortical connectivity.

Neural recording, known as local field potentials, offer valuable knowledge on how neural processes work and contribute to neural circuits. The recording can be contaminated by different internal and external sources of noise, because of the involvement of complex electronic apparatuses and the natural electrical activity throughout the body. In order to successfully utilize these signal, artifacts must be identified and removed. Thus, in this paper, an artifact detection method using a one-dimensional convolutional network referred to as 1D-CNN is proposed. The presented method achieved an improved accuracy and reduced computational time over the existing methods which use a multi-layered feed-forward neural network and a long-short term memory network. It also provided insight of the criteria behind the classification with gradient attribution maps.

The process of recording local fields potentials (LFP) can be contaminated by different internal and external sources of noise. To successfully use these recordings, noise must be removed, for which an automatic detection tool is needed to speed up the detection process. This work presents the use of a specific configuration of the recurrent neural network based machine learning approach, known as the long-short term memory (LSTM), in two different settings to identify artifacts and compares the obtained results to a feed forward neural network both in terms of classification performance and computational time. Using spontaneous LFP signals recorded chronically by multisite neuronal probes in behaving rats, our results show that the LSTM model with and without drop out can achieve an accuracy of 87.1%.

As our understanding of how motor output is generated increases, it is clear that there is a need to understand the interactions of multiple distinct regions rather than just the output properties of primary motor cortex. This becomes even more imperative when trying to understand how different regions may contribute to recovery following injury. In this study we used a technique that promotes functional motor recovery after injury, activity-dependent stimulation (ADS), to determine the short- and long-term effects on network activity and neuroplasticity of intracortical connections. ADS uses recorded neural activity to trigger stimulation of the brain and may be utilized to manipulate neuronal connectivity in vivo, representing a novel technique to shape intrinsic neuroplasticity. The aim of this work was to compare the effect of ADS to randomly-generated stimulation (RS) of the somatosensory area (S1) on the single units’ patterns of activity taking place in the premotor cortex (RFA) and to investigate whether synaptic plasticity changes occur in S1 as a consequence of 21 consecutive days of stimulation. In particular, we examined both firing rate changes and correlation between spiking activity and stimuli in chronically-implanted healthy ambulatory rats during both spontaneous and evoked activity, resulting from the two stimulation paradigms. Finally, we evaluated changes in expression of synaptophysin at the end of the treatment. This experimental procedure demonstrated the ability of ADS to modulate firing properties of RFA within daily recording sessions and to promote synaptogenesis in S1, further strengthening the idea that this Hebbian-inspired protocol can be used to effectively modulate cortical connectivity and thus suggesting its translational potential for promoting recovery after brain injury.

Background: Italy's first outbreak of COVID-19 was confirmed on 21 st February 2020;this virus has quickly become a global pandemic, causing a health emergen

One of the main limitations preventing the realization of a successful dialogue between the brain and a putative enabling device is the intricacy of brain signals. In this perspective, closed-loop in vitro systems can be used to investigate the interactions between a network of neurons and an external system, such as an interacting environment or an artificial device. In this chapter, we provide an overview of closed-loop in vitro systems, which have been developed for investigating potential neuroprosthetic applications. In particular, we first explore how to modify or set a target dynamical behavior in a network of neurons. We then analyze the behavior of in vitro systems connected to artificial devices, such as robots. Finally, we provide an overview of biological neuronal networks interacting with artificial neuronal networks, a configuration currently offering a promising solution for clinical applications.

Recent advances in neurotechnology allow neurological impairments to be treated or reduced by brain machine interfaces and neuroprostheses. To develop energy-efficient and real-time capable devices, neuromorphic computing systems are envisaged as the core of next-generation ‘neurobiohybrid’ systems for brain repair. We demonstrate here the first exploitation of a neuromorphic prosthesis to restore bidirectional interactions between two neuronal populations, even when one is damaged or completely missing. We used in vitro modular cell cultures to mimic the mutual interaction between neuronal assemblies and created a focal lesion to functionally disconnect the two populations. Then, we employed our neuromorphic prosthesis for two specific applications with future clinical implications: bidirectional bridging to artificially reconnect two disconnected neuronal modules and hybrid bidirectional bridging to replace the activity of one module with a neuromorphic spiking neural network. Our neuroprosthetic system opens up new avenues for the development of novel bioelectrical therapeutics for human applications.

One of the main limitations preventing the realization of a successful dialogue between the brain and a putative enabling device is the intricacy of brain signals. In this perspective, closed-loop in vitro systems can be used to investigate the interactions between a network of neurons and an external system, such as an interacting environment or an artificial device. In this chapter, we provide an overview of closed-loop in vitro systems, which have been developed for investigating potential neuroprosthetic applications. In particular, we first explore how to modify or set a target dynamical behavior in a network of neurons. We then analyze the behavior of in vitro systems connected to artificial devices, such as robots. Finally, we provide an overview of biological neuronal networks interacting with artificial neuronal networks, a configuration currently offering a promising solution for clinical applications.

BackgroundIntracortical microstimulation can be used successfully to modulate neuronal activity. Activity-dependent stimulation (ADS), in which action potentials recorded extracellularly from a single neuron are used to trigger stimulation at another cortical location (closed-loop), is an effective treatment for behavioral recovery after brain lesion in rodents. Neurophysiological changes in cortical communication induced by ADS, and how these changes differ from those induced by open-loop random stimulation (RS) are still not clear.ObjectivesWe investigated the ability of ADS and RS to induce changes in firing patterns in distant populations of neurons in healthy anesthetized rats.MethodsFor this study we used 23 adult Long-Evan rats, recording from a total of 591 neuronal units. Stimulation was delivered to either forelimb or barrel field somatosensory cortex, using either randomly-timed stimulus pulses or ADS triggered from neuronal spikes recorded in the rostral forelimb area (RFA) of the motor cortex.ResultsBoth RS and ADS stimulation protocols rapidly altered spike firing within RFA compared with no stimulation. Changes consisted of increases in mean firing rates and patterns of spike firing as measured by the revised Local Variation metric. ADS was more effective than RS in increasing short-latency evoked spikes during the stimulation periods, by producing a reliable, progressive increase in stimulus-related activity over time.ConclusionsThese results are critical for understanding the efficacy of electrical microstimulation protocols in altering activity patterns in interconnected brain networks. These data further strengthen the idea that activity-dependent microstimulation, can be used to modulate cortical state and functional connectivity.

Characterizing neuronal networks activity and their dynamical changes due to endogenous and exogenous causes is a key issue of computational neuroscience and constitutes a fundamental contribution towards the development of innovative intervention strategies in case of brain damage. We address this challenge by making use of a multimodular system able to confine the growth of cells on substrate-embedded microelectrode arrays to investigate the interactions between networks of neurons. We observed their spontaneous and electrically induced network activity before and after a laser cut disconnecting one of the modules from all the others. We found that laser dissection induced de-synchronized activity among different modules during spontaneous activity, and prevented the propagation of evoked responses among modules during electrical stimulation. This reproducible experimental model constitutes a test-bed for the design and development of innovative computational tools for characterizing neural damage, and of novel neuro-prostheses aimed at restoring lost neuronal functionality between distinct brain areas.

Intracortical microstimulation can be successfully used to manipulate neuronal activity and connectivity, thus representing a potentially powerful tool to steer neuroplasticity occurring after brain injury. Activity-dependent stimulation (ADS), in which the spikes recorded from a single neuron are used to trigger stimulation at another cortical location, is able to potentiate cortical connections between distant brain areas. Here, we developed an experimental procedure and a computational pipeline aimed at investigating the ability of ADS to induce changes in intra-cortical activity of healthy anesthetized rats.

The combination of recording and stimulation into closed-loop electrophysiological systems is one of the most challenging topics in neuroscience and neural engineering. The application of such systems can range from basic neuroscientific questions to neurorehabilitation purposes. In this review, we summarize the main results obtained with a hybrid neurorobotic platform that we developed in the past to perform closed-loop experiments. To minimize the complexity of the biological network, we used cultured neuronal networks kept alive over Micro Electrode Arrays. The robot is characterized by proximity sensors and wheels, allowing it to navigate within an arena with obstacles. The neurorobotic architecture can become an experimental environment for studying the interaction between brains and machines. We also discuss other possible applications of similar closed-loop architectures related to neuroprosthetics and neurorehabilitation.

This study assesses the feasibility of latent factor analysis via dynamical systems (LFADS) for evaluating differences in the observed spiking response dynamics imposed by two electrical microstimulation regimes in awake rats. LFADS is a recently-developed deep learning method that uses stimulus-aligned neural spiking data to determine the initial neural state of each trial, as well as infer a set of time-dependent perturbations to the learned neural dynamics within trials. We show that time-dependent perturbations inferred by an LFADS model trained on spikes from trials on a single session can distinguish between different stimulation conditions. Furthermore, we use these data to exemplify how LFADS inferences track the evolution of stimulus-related spiking responses during chronic microstimulation experiments.