Decoding Local Field Potentials for Neural Interfaces

68-channel neural signal processing system-on-chip with integrated feature extraction, compression, and hardware accelerators for neuroprosthetics in 22 nm FDSOI

Introduction

Multi-channel electrophysiology systems for recording of neuronal activity face significant data throughput limitations, hampering real-time, data-informed experiments. These limitations impact both experimental neurobiology research and next-generation neuroprosthetics.

Methods

We present a novel solution that leverages the high integration density of 22nm fully-depleted silicon-on-insulator technology to address these challenges. The proposed highly integrated programmable System-on-Chip (SoC) comprises 68-channel 0.41 μW/Ch recording frontends, spike detectors, 16-channel 0.87-4.39 μW/Ch action potentials and 8-channel 0.32 μW/Ch local field potential codecs, as well as a multiply-accumulate-assisted power-efficient processor operating at 25 MHz (5.19 μW/MHz). The system supports on-chip training processes for compression, training, and inference for neural spike sorting. The spike sorting achieves an average accuracy of 91.48 or 94.12% depending on the utilized features. The proposed programmable SoC is optimized for reduced area (9 mm2) and power. On-chip processing and compression capabilities free up the data bottlenecks in data transmission (up to 91% space saving ratio), and moreover enable a fully autonomous yet flexible processor-driven operation.

Discussion

Combined, these design considerations overcome data-bottlenecks by allowing on-chip feature extraction and subsequent compression.

Single-trial detection of auditory cues from the rat brain using memristors

Implantable devices hold the potential to address conditions currently lacking effective treatments, such as drug-resistant neural impairments and prosthetic control. Medical devices need to be biologically compatible while providing enhanced performance metrics of low-power consumption, high accuracy, small size, and minimal latency to enable ongoing intervention in brain function. Here, we demonstrate a memristor-based processing system for single-trial detection of behaviorally meaningful brain signals within a timeframe that supports real-time closed-loop intervention. We record neural activity from the reward center of the brain, the ventral tegmental area, in rats trained to associate a musical tone with a reward, and we use the memristors built-in thresholding properties to detect nontrivial biomarkers in local field potentials. This approach yields consistent and accurate detection of biomarkers >98% while maintaining power consumption as low as 4.14 nanowatt per channel. The efficacy of our system's capabilities to process real-time in vivo neural data paves the way for low-power chronic neural activity monitoring and biomedical implants.

Deep learning based decoding of single local field potential events

How is information processed in the cerebral cortex? In most cases, recorded brain activity is averaged over many (stimulus) repetitions, which erases the fine-structure of the neural signal. However, the brain is obviously a single-trial processor. Thus, we here demonstrate that an unsupervised machine learning approach can be used to extract meaningful information from electro-physiological recordings on a single-trial basis. We use an auto-encoder network to reduce the dimensions of single local field potential (LFP) events to create interpretable clusters of different neural activity patterns. Strikingly, certain LFP shapes correspond to latency differences in different recording channels. Hence, LFP shapes can be used to determine the direction of information flux in the cerebral cortex. Furthermore, after clustering, we decoded the cluster centroids to reverse-engineer the underlying prototypical LFP event shapes. To evaluate our approach, we applied it to both extra-cellular neural recordings in rodents, and intra-cranial EEG recordings in humans. Finally, we find that single channel LFP event shapes during spontaneous activity sample from the realm of possible stimulus evoked event shapes. A finding which so far has only been demonstrated for multi-channel population coding.

Beta-band power classification of go/no-go arm-reaching responses in the human hippocampus

Objective.Can we classify movement execution and inhibition from hippocampal oscillations during arm-reaching tasks? Traditionally associated with memory encoding, spatial navigation, and motor sequence consolidation, the hippocampus has come under scrutiny for its potential role in movement processing. Stereotactic electroencephalography (SEEG) has provided a unique opportunity to study the neurophysiology of the human hippocampus during motor tasks. In this study, we assess the accuracy of discriminant functions, in combination with principal component analysis (PCA), in classifying between 'Go' and 'No-go' trials in a Go/No-go arm-reaching task.Approach.Our approach centers on capturing the modulation of beta-band (13-30 Hz) power from multiple SEEG contacts in the hippocampus and minimizing the dimensional complexity of channels and frequency bins. This study utilizes SEEG data from the human hippocampus of 10 participants diagnosed with epilepsy. Spectral power was computed during a 'center-out' Go/No-go arm-reaching task, where participants reached or withheld their hand based on a colored cue. PCA was used to reduce data dimension and isolate the highest-variance components within the beta band. The Silhouette score was employed to measure the quality of clustering between 'Go' and 'No-go' trials. The accuracy of five different discriminant functions was evaluated using cross-validation.Main results.The Diagonal-Quadratic model performed best of the 5 classification models, exhibiting the lowest error rate in all participants (median: 9.91%, average: 14.67%). PCA showed that the first two principal components collectively accounted for 54.83% of the total variance explained on average across all participants, ranging from 36.92% to 81.25% among participants.Significance.This study shows that PCA paired with a Diagonal-Quadratic model can be an effective method for classifying between Go/No-go trials from beta-band power in the hippocampus during arm-reaching responses. This emphasizes the significance of hippocampal beta-power modulation in motor control, unveiling its potential implications for brain-computer interface applications.

Local contribution to the somatosensory evoked potentials in rat's thalamus

Local Field Potential (LFP), despite its name, often reflects remote activity. Depending on the orientation and synchrony of their sources, both oscillations and more complex waves may passively spread in brain tissue over long distances and be falsely interpreted as local activity at such distant recording sites. Here we show that the whisker-evoked potentials in the thalamic nuclei are of local origin up to around 6 ms post stimulus, but the later (7-15 ms) wave is overshadowed by a negative component reaching from cortex. This component can be analytically removed and local thalamic LFP can be recovered reliably using Current Source Density analysis. We used model-based kernel CSD (kCSD) method which allowed us to study the contribution of local and distant currents to LFP from rat thalamic nuclei and barrel cortex recorded with multiple, non-linear and non-regular multichannel probes. Importantly, we verified that concurrent recordings from the cortex are not essential for reliable thalamic CSD estimation. The proposed framework can be used to analyze LFP from other brain areas and has consequences for general LFP interpretation and analysis.

Machine learning-based high-frequency neuronal spike reconstruction from low-frequency and low-sampling-rate recordings

Recording neuronal activity using multiple electrodes has been widely used to understand the functional mechanisms of the brain. Increasing the number of electrodes allows us to decode more variety of functionalities. However, handling massive amounts of multichannel electrophysiological data is still challenging due to the limited hardware resources and unavoidable thermal tissue damage. Here, we present machine learning (ML)-based reconstruction of high-frequency neuronal spikes from subsampled low-frequency band signals. Inspired by the equivalence between high-frequency restoration and super-resolution in image processing, we applied a transformer ML model to neuronal data recorded from both in vitro cultures and in vivo male mouse brains. Even with the x8 downsampled datasets, our trained model reasonably estimated high-frequency information of spiking activity, including spike timing, waveform, and network connectivity. With our ML-based data reduction applicable to existing multichannel recording hardware while achieving neuronal signals of broad bandwidths, we expect to enable more comprehensive analysis and control of brain functions.

A Hyperflexible Electrode Array for Long-Term Recording and Decoding of Intraspinal Neuronal Activity

Neural interfaces for stable access to the spinal cord (SC) electrical activity can benefit patients with motor dysfunctions. Invasive high-density electrodes can directly extract signals from SC neuronal populations that can be used for the facilitation, adjustment, and reconstruction of motor actions. However, developing neural interfaces that can achieve high channel counts and long-term intraspinal recording remains technically challenging. Here, a biocompatible SC hyperflexible electrode array (SHEA) with an ultrathin structure that minimizes mechanical mismatch between the interface and SC tissue and enables stable single-unit recording for more than 2 months in mice is demonstrated. These results show that SHEA maintains stable impedance, signal-to-noise ratio, single-unit yield, and spike amplitude after implantation into mouse SC. Gait analysis and histology show that SHEA implantation induces negligible behavioral effects and Inflammation. Additionally, multi-unit signals recorded from the SC ventral horn can predict the mouse's movement trajectory with a high decoding coefficient of up to 0.95. Moreover, during step cycles, it is found that the neural trajectory of spikes and low-frequency local field potential (LFP) signal exhibits periodic geometry patterns. Thus, SHEA can offer an efficient and reliable SC neural interface for monitoring and potentially modulating SC neuronal activity associated with motor dysfunctions.

Neural Decoding for Intracortical Brain-Computer Interfaces

Brain-computer interfaces have revolutionized the field of neuroscience by providing a solution for paralyzed patients to control external devices and improve the quality of daily life. To accurately and stably control effectors, it is important for decoders to recognize an individual's motor intention from neural activity either by noninvasive or intracortical neural recording. Intracortical recording is an invasive way of measuring neural electrical activity with high temporal and spatial resolution. Herein, we review recent developments in neural signal decoding methods for intracortical brain-computer interfaces. These methods have achieved good performance in analyzing neural activity and controlling robots and prostheses in nonhuman primates and humans. For more complex paradigms in motor rehabilitation or other clinical applications, there remains more space for further improvements of decoders.

Representation and decoding of bilateral arm motor imagery using unilateral cerebral LFP signals

Introduction

In the field of upper limb brain computer interfaces (BCIs), the research focusing on bilateral decoding mostly based on the neural signals from two cerebral hemispheres. In addition, most studies used spikes for decoding. Here we examined the representation and decoding of different laterality and regions arm motor imagery in unilateral motor cortex based on local field potentials (LFPs).

Methods

The LFP signals were recorded from a 96-channel Utah microelectrode array implanted in the left primary motor cortex of a paralyzed participant. There were 7 kinds of tasks: rest, left, right and bilateral elbow and wrist flexion. We performed time-frequency analysis on the LFP signals and analyzed the representation and decoding of different tasks using the power and energy of different frequency bands.

Results

The frequency range of <8 Hz and >38 Hz showed power enhancement, whereas 8-38 Hz showed power suppression in spectrograms while performing motor imagery. There were significant differences in average energy between tasks. What's more, the movement region and laterality were represented in two dimensions by demixed principal component analysis. The 135-300 Hz band signal had the highest decoding accuracy among all frequency bands and the contralateral and bilateral signals had more similar single-channel power activation patterns and larger signal correlation than contralateral and ipsilateral signals, bilateral and ipsilateral signals.

Discussion

The results showed that unilateral LFP signals had different representations for bilateral motor imagery on the average energy of the full array and single-channel power levels, and different tasks could be decoded. These proved the feasibility of multilateral BCI based on the unilateral LFP signal to broaden the application of BCI technology.

Clinical trial registration

https://www.chictr.org.cn/showproj.aspx?proj=130829, identifier ChiCTR2100050705.

Long-term unsupervised recalibration of cursor BCIs

Intracortical brain-computer interfaces (iBCIs) require frequent recalibration to maintain robust performance due to changes in neural activity that accumulate over time. Compensating for this nonstationarity would enable consistently high performance without the need for supervised recalibration periods, where users cannot engage in free use of their device. Here we introduce a hidden Markov model (HMM) to infer what targets users are moving toward during iBCI use. We then retrain the system using these inferred targets, enabling unsupervised adaptation to changing neural activity. Our approach outperforms the state of the art in large-scale, closed-loop simulations over two months and in closed-loop with a human iBCI user over one month. Leveraging an offline dataset spanning five years of iBCI recordings, we further show how recently proposed data distribution-matching approaches to recalibration fail over long time scales; only target-inference methods appear capable of enabling long-term unsupervised recalibration. Our results demonstrate how task structure can be used to bootstrap a noisy decoder into a highly-performant one, thereby overcoming one of the major barriers to clinically translating BCIs.

Classification of Whisker Deflections From Evoked Responses in the Somatosensory Barrel Cortex With Spiking Neural Networks

Spike-based neuromorphic hardware has great potential for low-energy brain-machine interfaces, leading to a novel paradigm for neuroprosthetics where spiking neurons in silicon read out and control activity of brain circuits. Neuromorphic processors can receive rich information about brain activity from both spikes and local field potentials (LFPs) recorded by implanted neural probes. However, it was unclear whether spiking neural networks (SNNs) implemented on such devices can effectively process that information. Here, we demonstrate that SNNs can be trained to classify whisker deflections of different amplitudes from evoked responses in a single barrel of the rat somatosensory cortex. We show that the classification performance is comparable or even superior to state-of-the-art machine learning approaches. We find that SNNs are rather insensitive to recorded signal type: both multi-unit spiking activity and LFPs yield similar results, where LFPs from cortical layers III and IV seem better suited than those of deep layers. In addition, no hand-crafted features need to be extracted from the data—multi-unit activity can directly be fed into these networks and a simple event-encoding of LFPs is sufficient for good performance. Furthermore, we find that the performance of SNNs is insensitive to the network state—their performance is similar during UP and DOWN states.

Decoding Bilateral Hindlimb Kinematics From Cat Spinal Signals Using Three-Dimensional Convolutional Neural Network

To date, decoding limb kinematic information mostly relies on neural signals recorded from the peripheral nerve, dorsal root ganglia (DRG), ventral roots, spinal cord gray matter, and the sensorimotor cortex. In the current study, we demonstrated that the neural signals recorded from the lateral and dorsal columns within the spinal cord have the potential to decode hindlimb kinematics during locomotion. Experiments were conducted using intact cats. The cats were trained to walk on a moving belt in a hindlimb-only condition, while their forelimbs were kept on the front body of the treadmill. The bilateral hindlimb joint angles were decoded using local field potential signals recorded using a microelectrode array implanted in the dorsal and lateral columns of both the left and right sides of the cat spinal cord. The results show that contralateral hindlimb kinematics can be decoded as accurately as ipsilateral kinematics. Interestingly, hindlimb kinematics of both legs can be accurately decoded from the lateral columns within one side of the spinal cord during hindlimb-only locomotion. The results indicated that there was no significant difference between the decoding performances obtained using neural signals recorded from the dorsal and lateral columns. The results of the time-frequency analysis show that event-related synchronization (ERS) and event-related desynchronization (ERD) patterns in all frequency bands could reveal the dynamics of the neural signals during movement. The onset and offset of the movement can be clearly identified by the ERD/ERS patterns. The results of the mutual information (MI) analysis showed that the theta frequency band contained significantly more limb kinematics information than the other frequency bands. Moreover, the theta power increased with a higher locomotion speed.

CluSem: Accurate clustering-based ensemble method to predict motor imagery tasks from multi-channel EEG data

BACKGROUND

The classification of motor imagery electroencephalogram (MI-EEG) is a pivotal task in the biosignal classification process in the brain-computer interface (BCI) applications. Currently, this bio-engineering-based technology is being employed by researchers in various fields to develop cutting-edge applications. The classification of real-time MI-EEG signals is the most challenging task in these applications. The prediction performance of the existing classification methods is still limited due to the high dimensionality and dynamic behaviors of the real-time EEG data.

PROPOSED METHOD

To enhance the classification performance of real-time BCI applications, this paper presents a new clustering-based ensemble technique called CluSem to mitigate this problem. We also develop a new brain game called CluGame using this method to evaluate the classification performance of real-time motor imagery movements. In this game, real-time EEG signal classification and prediction tabulation through animated balls are controlled via threads. By playing this game, users can control the movements of the balls via the brain signals of motor imagery movements without using any traditional input devices.

RESULTS

Our results demonstrate that CluSem is able to improve the classification accuracy between 5% and 15% compared to the existing methods on our collected as well as the publicly available EEG datasets. The source codes used to implement CluSem and CluGame are publicly available at https://github.com/MdOchiuddinMiah/MI-BCI_ML.

Implantable brain machine interfaces: first-in-human studies, technology challenges and trends

Implantable brain machine interfaces (BMIs) are now on a trajectory to go mainstream, wherein what was once considered last resort will progressively become elective at earlier stages in disease treatment. First-in-human successes have demonstrated the ability to decode highly dexterous motor skills such as handwriting, and speech from human cortical activity. These have been used for cursor and prosthesis control, direct-to-text communication and speech synthesis. Along with these breakthrough studies, technology advancements have enabled the observation of more channels of neural activity through new concepts for centralised/distributed implant architectures. This is complemented by research in flexible substrates, packaging, surgical workflows and data processing. New regulatory guidance and funding has galvanised the field. This culmination of resource, efforts and capability is now attracting significant investment for BMI commercialisation. This paper reviews recent developments and describes the paradigm shift in BMI development that is leading to new innovations, insights and BMI translation.

A Brain Biometric-based Identification Approach Using Local Field Potentials

Traditional biometrics such as face, iris and fingerprint have been applied widely nowadays. Nevertheless, with more and more potential problems being exposed, such as privacy leak and fabricate attack, it is urgent to find new secure biometrics to meet the needs. Identification based on brain signals is a promising option due to its unique advantages of confidentiality, anti-spoofing, continuity and cancelability. Among various types of brain signals, local field potential (LFP) has long term stability, high signal to noise ratio and high spatial resolution, which is suitable for identification. In this paper, we propose a novel biometric which is extracted from LFP signals with a deep neural network. The proposed biometric can be generated in a task-related manner thus is cancelable. Experiments with ten rats demonstrate that, the proposed biometric achieves a high identification accuracy of 94.47%, and the performance is stable over several days.

Inferring entire spiking activity from local field potentials

Extracellular recordings are typically analysed by separating them into two distinct signals: local field potentials (LFPs) and spikes. Previous studies have shown that spikes, in the form of single-unit activity (SUA) or multiunit activity (MUA), can be inferred solely from LFPs with moderately good accuracy. SUA and MUA are typically extracted via threshold-based technique which may not be reliable when the recordings exhibit a low signal-to-noise ratio (SNR). Another type of spiking activity, referred to as entire spiking activity (ESA), can be extracted by a threshold-less, fast, and automated technique and has led to better performance in several tasks. However, its relationship with the LFPs has not been investigated. In this study, we aim to address this issue by inferring ESA from LFPs intracortically recorded from the motor cortex area of three monkeys performing different tasks. Results from long-term recording sessions and across subjects revealed that ESA can be inferred from LFPs with good accuracy. On average, the inference performance of ESA was consistently and significantly higher than those of SUA and MUA. In addition, local motor potential (LMP) was found to be the most predictive feature. The overall results indicate that LFPs contain substantial information about spiking activity, particularly ESA. This could be useful for understanding LFP-spike relationship and for the development of LFP-based BMIs.

Brain-Machine Interfaces: Lessons for Prosthetic Hand Control

Brain-machine interfaces (BMI) are being developed to restore upper limb function for persons with spinal cord injury or other motor degenerative conditions. BMI and implantable sensors for myoelectric prostheses directly extract information from the central or peripheral nervous system to provide users with high fidelity control of their prosthetic device. Control algorithms have been highly transferable between the 2 technologies but also face common issues. In this review of the current state of the art in each field, the authors point out similarities and differences between the 2 technologies that may guide the implementation of common solutions to these challenges.

Brain-Computer-Spinal Interface Restores Upper Limb Function After Spinal Cord Injury

Brain-computer interfaces (BCIs) are an emerging strategy for spinal cord injury (SCI) intervention that may be used to reanimate paralyzed limbs. This approach requires decoding movement intention from the brain to control movement-evoking stimulation. Common decoding methods use spike-sorting and require frequent calibration and high computational complexity. Furthermore, most applications of closed-loop stimulation act on peripheral nerves or muscles, resulting in rapid muscle fatigue. Here we show that a local field potential-based BCI can control spinal stimulation and improve forelimb function in rats with cervical SCI. We decoded forelimb movement via multi-channel local field potentials in the sensorimotor cortex using a canonical correlation analysis algorithm. We then used this decoded signal to trigger epidural spinal stimulation and restore forelimb movement. Finally, we implemented this closed-loop algorithm in a miniaturized onboard computing platform. This Brain-Computer-Spinal Interface (BCSI) utilized recording and stimulation approaches already used in separate human applications. Our goal was to demonstrate a potential neuroprosthetic intervention to improve function after upper extremity paralysis.

Deep Pinsker and James-Stein Neural Networks for Decoding Motor Intentions From Limited Data

Non-parametric regression has been shown to be useful in extracting relevant features from Local Field Potential (LFP) signals for decoding motor intentions. Yet, in many instances, brain-computer interfaces (BCIs) rely on simple classification methods, circumventing deep neural networks (DNNs) due to limited training data. This paper leverages the robustness of several important results in non-parametric regression to harness the potentials of deep learning in limited data setups. We consider a solution that combines Pinsker's theorem as well as its adaptively optimal counterpart due to James-Stein for feature extraction from LFPs, followed by a DNN for classifying motor intentions. We apply our approach to the problem of decoding eye movement intentions from LFPs collected in macaque cortex while the animals perform memory-guided visual saccades to one of eight target locations. The results demonstrate that a DNN classifier trained over the Pinsker features outperforms the benchmark method based on linear discriminant analysis (LDA) trained over the same features.

Decoding route selection of pigeon during goal-directed behavior: A joint spike-LFP study

How to reach the goal is one of the core problems that animals must solve to complete goal-directed behavior. Studies have proved the important role of hippocampus (Hp) in spatial navigation and shown that hippocampal neural activities can represent the current location and goal location. However, for the different routes linking these two locations, the neural representation mechanism of the route selection in Hp is not clear. Here, we addressed this question using neural recordings of Hp ensembles and decoding analyses in pigeons performing a goal-directed route selection task known to require Hp participation. The hippocampal spike trains and local field potentials (LFPs) of five pigeons performing the task were acquired and analyzed. We found that the neuron firing rates and power spectrum characteristics in Hp could encode the animal's route selection during goal-directed behavior, suggesting that the representation of route selection was coherent for hippocampal spike and LFP signals. Decoding results further indicated that joint spike-LFP features resulted in a significant improvement in the representation accuracy of the route selection. These findings of this study will help to understand the encoding mechanism of route selection in goal-directed behavior.

Force Decoding of Caudal Forelimb Area and Rostral Forelimb Area in Chronic Stroke Rats

Brain machine interfaces (BMIs) used for movement restoration primarily rely on studies of motor decoding. It has been proved that local field potentials (LFPs) from primary motor cortex and premotor cortex of normal rodents could be used for decoding motor signals. However, few studies have explored the decoding performance of these brain areas under motor cortex damage. In this work, we focus on force decoding performance of LFPs spectrum from both ipsilesional caudal forelimb area (CFA) and rostral forelimb area (RFA) of rodents with ischemia over CFA. After three months of ischemia induced by photothrombosis over CFA, the power of high-frequency bands (>120 Hz) from both CFA and RFA can decode force signals by Kalman filters. The fair performance of CFA indicates motor reorganization over penumbra. Further exploration of RFA decoding ability proves that at least four electrodes of RFA should be used on decoding and electrodes far from CFA of stroke rats could achieve almost as good results as those close to CFA of normal rats, which indicates the motor remapping. Experimental results show the long-term stability of PM LFPs decoding performance of stroke rats as the trained Kalman model could be used to accurately decode force some days later which provides a possibility for online decoding system. In conclusion, our work shows that even under CFA ischemia, high-frequency power of LFPs from RFA is still able to accurately decode force signals and has long stability, which provides the possibility of BMIs for motor function reconstruction of chronic stroke patients.

Robust and accurate decoding of hand kinematics from entire spiking activity using deep learning

OBJECTIVE

Brain-machine interfaces (BMIs) seek to restore lost motor functions in individuals with neurological disorders by enabling them to control external devices directly with their thoughts. This work aims to improve robustness and decoding accuracy that currently become major challenges in the clinical translation of intracortical BMIs.

APPROACH

We propose entire spiking activity (ESA) -an envelope of spiking activity that can be extracted by a simple, threshold-less, and automated technique- as the input signal. We couple ESA with deep learning-based decoding algorithm that uses quasi-recurrent neural network (QRNN) architecture. We evaluate comprehensively the performance of ESA-driven QRNN decoder for decoding hand kinematics from neural signals chronically recorded from the primary motor cortex area of three non-human primates performing different tasks.

MAIN RESULTS

Our proposed method yields consistently higher decoding performance than any other combinations of the input signal and decoding algorithm previously reported across long term recording sessions. It can sustain high decoding performance even when removing spikes from the raw signals, when using the different number of channels, and when using a smaller amount of training data.

SIGNIFICANCE

Overall results demonstrate exceptionally high decoding accuracy and chronic robustness, which is highly desirable given it is an unresolved challenge in BMIs.

Generalized neural decoders for transfer learning across participants and recording modalities

OBJECTIVE

Advances in neural decoding have enabled brain-computer interfaces to perform increasingly complex and clinically-relevant tasks. However, such decoders are often tailored to specific participants, days, and recording sites, limiting their practical long-term usage. Therefore, a fundamental challenge is to develop neural decoders that can robustly train on pooled, multi-participant data and generalize to new participants.

APPROACH

We introduce a new decoder, HTNet, which uses a convolutional neural network with two innovations: (1) a Hilbert transform that computes spectral power at data-driven frequencies and (2) a layer that projects electrode-level data onto predefined brain regions. The projection layer critically enables applications with intracranial electrocorticography (ECoG), where electrode locations are not standardized and vary widely across participants. We trained HTNet to decode arm movements using pooled ECoG data from 11 of 12 participants and tested performance on unseen ECoG or electroencephalography (EEG) participants; these pretrained models were also subsequently fine-tuned to each test participant.

MAIN RESULTS

HTNet outperformed state-of-the-art decoders when tested on unseen participants, even when a different recording modality was used. By fine-tuning these generalized HTNet decoders, we achieved performance approaching the best tailored decoders with as few as 50 ECoG or 20 EEG events. We were also able to interpret HTNet's trained weights and demonstrate its ability to extract physiologically-relevant features.

SIGNIFICANCE

By generalizing to new participants and recording modalities, robustly handling variations in electrode placement, and allowing participant-specific fine-tuning with minimal data, HTNet is applicable across a broader range of neural decoding applications compared to current state-of-the-art decoders.

Impact of referencing scheme on decoding performance of LFP-based brain-machine interface

OBJECTIVE

There has recently been an increasing interest in local field potential (LFP) for brain-machine interface (BMI) applications due to its desirable properties (signal stability and low bandwidth). LFP is typically recorded with respect to a single unipolar reference which is susceptible to common noise. Several referencing schemes have been proposed to eliminate the common noise, such as bipolar reference, current source density (CSD), and common average reference (CAR). However, to date, there have not been any studies to investigate the impact of these referencing schemes on decoding performance of LFP-based BMIs.

APPROACH

To address this issue, we comprehensively examined the impact of different referencing schemes and LFP features on the performance of hand kinematic decoding using a deep learning method. We used LFPs chronically recorded from the motor cortex area of a monkey while performing reaching tasks.

MAIN RESULTS

Experimental results revealed that local motor potential (LMP) emerged as the most informative feature regardless of the referencing schemes. Using LMP as the feature, CAR was found to yield consistently better decoding performance than other referencing schemes over long-term recording sessions. Significance Overall, our results suggest the potential use of LMP coupled with CAR for enhancing the decoding performance of LFP-based BMIs.

Reinforcement Learning Based Fast Self-Recalibrating Decoder for Intracortical Brain-Machine Interface

BACKGROUND

For the nonstationarity of neural recordings in intracortical brain-machine interfaces, daily retraining in a supervised manner is always required to maintain the performance of the decoder. This problem can be improved by using a reinforcement learning (RL) based self-recalibrating decoder. However, quickly exploring new knowledge while maintaining a good performance remains a challenge in RL-based decoders.

METHODS

To solve this problem, we proposed an attention-gated RL-based algorithm combining transfer learning, mini-batch, and weight updating schemes to accelerate the weight updating and avoid over-fitting. The proposed algorithm was tested on intracortical neural data recorded from two monkeys to decode their reaching positions and grasping gestures.

RESULTS

The decoding results showed that our proposed algorithm achieved an approximate 20% increase in classification accuracy compared to that obtained by the non-retrained classifier and even achieved better classification accuracy than the daily retraining classifier. Moreover, compared with a conventional RL method, our algorithm improved the accuracy by approximately 10% and the online weight updating speed by approximately 70 times.

CONCLUSIONS

This paper proposed a self-recalibrating decoder which achieved a good and robust decoding performance with fast weight updating and might facilitate its application in wearable device and clinical practice.

Agency and Accountability: Ethical Considerations for Brain-Computer Interfaces

Brain-computer interfaces (BCIs) are systems in which a user's real-time brain activity is used to control an external device, such as a prosthetic limb. BCIs have great potential for restoring lost motor functions in a wide range of patients. However, this futuristic technology raises several ethical questions, especially concerning the degree of agency a BCI affords its user and the extent to which a BCI user ought to be accountable for actions undertaken via the device. This paper examines these and other ethical concerns found at each of the three major parts of the BCI system: the sensor that records neural activity, the decoder that converts raw data into usable signals, and the translator that uses these signals to control the movement of an external device.

Untangling Neural Representations in the Motor Cortex

How the brain generates accurate movement is a long-standing problem in neuroscience. In this issue of Neuron, Russo et al. (2018) argue that population activity in motor cortex does not represent muscle patterns but rather untangled neural trajectories that are robust to noise.

Minimax-optimal decoding of movement goals from local field potentials using complex spectral features

OBJECTIVE

We consider the problem of predicting eye movement goals from local field potentials (LFP) recorded through a multielectrode array in the macaque prefrontal cortex. The monkey is tasked with performing memory-guided saccades to one of eight targets during which LFP activity is recorded and used to train a decoder.

APPROACH

Previous reports have mainly relied on the spectral amplitude of the LFPs as decoding feature, while neglecting the phase without proper theoretical justification. This paper formulates the problem of decoding eye movement intentions in a statistically optimal framework and uses Gaussian sequence modeling and Pinsker's theorem to generate minimax-optimal estimates of the LFP signals which are used as decoding features. The approach is shown to act as a low-pass filter and each LFP in the feature space is represented via its complex Fourier coefficients after appropriate shrinking such that higher frequency components are attenuated; this way, the phase information inherently present in the LFP signal is naturally embedded into the feature space.

MAIN RESULTS

We show that the proposed complex spectrum-based decoder achieves prediction accuracy of up to [Formula: see text] at superficial cortical depths near the surface of the prefrontal cortex; this marks a significant performance improvement over conventional power spectrum-based decoders.

SIGNIFICANCE

The presented analyses showcase the promising potential of low-pass filtered LFP signals for highly reliable neural decoding of intended motor actions.

Brain-Machine Interfaces: Powerful Tools for Clinical Treatment and Neuroscientific Investigations

Brain-machine interfaces (BMIs) have exploded in popularity in the past decade. BMIs, also called brain-computer interfaces, provide a direct link between the brain and a computer, usually to control an external device. BMIs have a wide array of potential clinical applications, ranging from restoring communication to people unable to speak due to amyotrophic lateral sclerosis or a stroke, to restoring movement to people with paralysis from spinal cord injury or motor neuron disease, to restoring memory to people with cognitive impairment. Because BMIs are controlled directly by the activity of prespecified neurons or cortical areas, they also provide a powerful paradigm with which to investigate fundamental questions about brain physiology, including neuronal behavior, learning, and the role of oscillations. This article reviews the clinical and neuroscientific applications of BMIs, with a primary focus on motor BMIs.

Long-term recording reliability of liquid crystal polymer µECoG arrays

OBJECTIVE

The clinical use of microsignals recorded over broad cortical regions is largely limited by the chronic reliability of the implanted interfaces.

APPROACH

We evaluated the chronic reliability of novel 61-channel micro-electrocorticographic (µECoG) arrays in rats chronically implanted for over one year and using accelerated aging. Devices were encapsulated with polyimide (PI) or liquid crystal polymer (LCP), and fabricated using commercial manufacturing processes. In vitro failure modes and predicted lifetimes were determined from accelerated soak testing. Successful designs were implanted epidurally over the rodent auditory cortex. Trends in baseline signal level, evoked responses and decoding performance were reported for over one year of implantation.

MAIN RESULTS

Devices fabricated with LCP consistently had longer in vitro lifetimes than PI encapsulation. Our accelerated aging results predicted device integrity beyond 3.4 years. Five implanted arrays showed stable performance over the entire implantation period (247-435 d). Our regression analysis showed that impedance predicted signal quality and information content only in the first 31 d of recordings and had little predictive value in the chronic phase (>31 d). In the chronic phase, site impedances slightly decreased yet decoding performance became statistically uncorrelated with impedance. We also employed an improved statistical model of spatial variation to measure sensitivity to locally varying fields, which is typically concealed in standard signal power calculations.

SIGNIFICANCE

These findings show that µECoG arrays can reliably perform in chronic applications in vivo for over one year, which facilitates the development of a high-density, clinically viable interface.

h-Type Membrane Current Shapes the Local Field Potential from Populations of Pyramidal Neurons

In cortex, the local field potential (LFP) is thought to mainly stem from correlated synaptic input to populations of geometrically aligned neurons. Computer models of single cortical pyramidal neurons showed that subthreshold voltage-dependent membrane conductances can also shape the LFP signal, in particular the hyperpolarization-activated cation current (I_h; h-type). This ion channel is prominent in various types of pyramidal neurons, typically showing an increasing density gradient along the apical dendrites. Here, we investigate how I_h affects the LFP generated by a model of a population of cortical pyramidal neurons. We find that the LFP from populations of neurons that receive uncorrelated synaptic input can be well predicted by the LFP from single neurons. In this case, when input impinges on the distal dendrites, where most h-type channels are located, a strong resonance in the LFP was measured near the soma, whereas the opposite configuration does not reveal an I_h contribution to the LFP. Introducing correlations in the synaptic inputs to the pyramidal cells strongly amplifies the LFP, while maintaining the differential effects of I_h for distal dendritic versus perisomatic input. Previous theoretical work showed that input correlations do not amplify LFP power when neurons receive synaptic input uniformly across the cell. We find that this crucially depends on the membrane conductance distribution: the asymmetric distribution of I_h results in a strong amplification of the LFP when synaptic inputs to the cell population are correlated. In conclusion, we find that the h-type current is particularly suited to shape the LFP signal in cortical populations.SIGNIFICANCE STATEMENT The local field potential (LFP), the low-frequency part of extracellular potentials recorded in neural tissue, is often used for probing neural circuit activity. While the cortical LFP is thought to mainly reflect synaptic inputs onto pyramidal neurons, little is known about the role of subthreshold active conductances in shaping the LFP. By means of biophysical modeling we obtain a comprehensive, qualitative understanding of how LFPs generated by populations of cortical pyramidal neurons depend on active subthreshold currents, and identify the key importance of the h-type channel. Our results show that LFPs can give information about the active properties of neurons and that preferred frequencies in the LFP can result from those cellular properties instead of, for example, network dynamics.

On-Probe Neural Interface ASIC for Combined Electrical Recording and Optogenetic Stimulation

Neuromodulation technologies are progressing from pacemaking and sensory operations to full closed-loop control. In particular, optogenetics-the genetic modification of light sensitivity into neural tissue allows for simultaneous optical stimulation and electronic recording. This paper presents a neural interface application-specified integrated circuit (ASIC) for intelligent optoelectronic probes. The architecture is designed to enable simultaneous optical neural stimulation and electronic recording. It provides four low noise (2.08  μV) recording channels optimized for recording local field potentials (LFPs) (0.1-300 Hz bandwidth, 5 mV range, sampled 10-bit@4 kHz), which are more stable for chronic applications. For stimulation, it provides six independently addressable optical driver circuits, which can provide both intensity (8-bit resolution across a 1.1 mA range) and pulse-width modulation for high-radiance light emitting diodes (LEDs). The system includes a fully digital interface using a serial peripheral interface (SPI) protocol to allow for use with embedded controllers. The SPI interface is embedded within a finite state machine (FSM), which implements a command interpreter that can send out LFP data whilst receiving instructions to control LED emission. The circuit has been implemented in a commercially available 0.35  μm CMOS technology occupying a 1.95 mm 1.10 mm footprint for mounting onto the head of a silicon probe. Measured results are given for a variety of bench-top, in vitro and in vivo experiments, quantifying system performance and also demonstrating concurrent recording and stimulation within relevant experimental models.

Physiological properties of brain-machine interface input signals

Brain-machine interfaces (BMIs), also called brain-computer interfaces (BCIs), decode neural signals and use them to control some type of external device. Despite many experimental successes and terrific demonstrations in animals and humans, a high-performance, clinically viable device has not yet been developed for widespread usage. There are many factors that impact clinical viability and BMI performance. Arguably, the first of these is the selection of brain signals used to control BMIs. In this review, we summarize the physiological characteristics and performance-including movement-related information, longevity, and stability-of multiple types of input signals that have been used in invasive BMIs to date. These include intracortical spikes as well as field potentials obtained inside the cortex, at the surface of the cortex (electrocorticography), and at the surface of the dura mater (epidural signals). We also discuss the potential for future enhancements in input signal performance, both by improving hardware and by leveraging the knowledge of the physiological characteristics of these signals to improve decoding and stability.