Demonstration of a portable intracortical brain-computer interface

An emerging view of neural geometry in motor cortex supports high-performance decoding

Decoders for brain-computer interfaces (BCIs) assume constraints on neural activity, chosen to reflect scientific beliefs while yielding tractable computations. Recent scientific advances suggest that the true constraints on neural activity, especially its geometry, may be quite different from those assumed by most decoders. We designed a decoder, MINT, to embrace statistical constraints that are potentially more appropriate. If those constraints are accurate, MINT should outperform standard methods that explicitly make different assumptions. Additionally, MINT should be competitive with expressive machine learning methods that can implicitly learn constraints from data. MINT performed well across tasks, suggesting its assumptions are well-matched to the data. MINT outperformed other interpretable methods in every comparison we made. MINT outperformed expressive machine learning methods in 37 of 42 comparisons. MINT's computations are simple, scale favorably with increasing neuron counts, and yield interpretable quantities such as data likelihoods. MINT's performance and simplicity suggest it may be a strong candidate for many BCI applications.

Enhancing neuroprosthesis calibration: the advantage of integrating prior training over exclusive use of new data

Objective.Neuroprostheses typically operate under supervised learning, in which a machine-learning algorithm is trained to correlate neural or myoelectric activity with an individual's motor intent. Due to the stochastic nature of neuromyoelectric signals, algorithm performance decays over time. This decay is accelerated when attempting to regress proportional control of multiple joints in parallel, compared with the more typical classification-based pattern recognition control. To overcome this degradation, neuroprostheses and commercial myoelectric prostheses are often recalibrated and retrained frequently so that only the most recent, up-to-date data influences the algorithm performance. Here, we introduce and validate an alternative training paradigm in which training data from past calibrations is aggregated and reused in future calibrations for regression control.Approach.Using a cohort of four transradial amputees implanted with intramuscular electromyographic recording leads, we demonstrate that aggregating prior datasets improves prosthetic regression-based control in offline analyses and an online human-in-the-loop task. In offline analyses, we compared the performance of a convolutional neural network (CNN) and a modified Kalman filter (MKF) to simultaneously regress the kinematics of an eight-degree-of-freedom prosthesis. Both algorithms were trained under the traditional paradigm using a single dataset, as well as under the new paradigm using aggregated datasets from the past five or ten trainings.Main results.Dataset aggregation reduced the root-mean-squared error (RMSE) of algorithm estimates for both the CNN and MKF, although the CNN saw a greater reduction in error. Further offline analyses revealed that dataset aggregation improved CNN robustness when reusing the same algorithm on subsequent test days, as indicated by a smaller increase in RMSE per day. Finally, data from an online virtual-target-touching task with one amputee showed significantly better real-time prosthetic control when using aggregated training data from just two prior datasets.Significance.Altogether, these results demonstrate that training data from past calibrations should not be discarded but, rather, should be reused in an aggregated training dataset such that the increased amount and diversity of data improve algorithm performance. More broadly, this work supports a paradigm shift for the field of neuroprostheses away from daily data recalibration for linear classification models and towards daily data aggregation for non-linear regression models.

Speech motor cortex enables BCI cursor control and click

Decoding neural activity from ventral (speech) motor cortex is known to enable high-performance speech brain-computer interface (BCI) control. It was previously unknown whether this brain area could also enable computer control via neural cursor and click, as is typically associated with dorsal (arm and hand) motor cortex. We recruited a clinical trial participant with ALS and implanted intracortical microelectrode arrays in ventral precentral gyrus (vPCG), which the participant used to operate a speech BCI in a prior study. We developed a cursor BCI driven by the participant's vPCG neural activity, and evaluated performance on a series of target selection tasks. The reported vPCG cursor BCI enabled rapidly-calibrating (40 seconds), accurate (2.90 bits per second) cursor control and click. The participant also used the BCI to control his own personal computer independently. These results suggest that placing electrodes in vPCG to optimize for speech decoding may also be a viable strategy for building a multi-modal BCI which enables both speech-based communication and computer control via cursor and click.

The speech neuroprosthesis

Loss of speech after paralysis is devastating, but circumventing motor-pathway injury by directly decoding speech from intact cortical activity has the potential to restore natural communication and self-expression. Recent discoveries have defined how key features of speech production are facilitated by the coordinated activity of vocal-tract articulatory and motor-planning cortical representations. In this Review, we highlight such progress and how it has led to successful speech decoding, first in individuals implanted with intracranial electrodes for clinical epilepsy monitoring and subsequently in individuals with paralysis as part of early feasibility clinical trials to restore speech. We discuss high-spatiotemporal-resolution neural interfaces and the adaptation of state-of-the-art speech computational algorithms that have driven rapid and substantial progress in decoding neural activity into text, audible speech, and facial movements. Although restoring natural speech is a long-term goal, speech neuroprostheses already have performance levels that surpass communication rates offered by current assistive-communication technology. Given this accelerated rate of progress in the field, we propose key evaluation metrics for speed and accuracy, among others, to help standardize across studies. We finish by highlighting several directions to more fully explore the multidimensional feature space of speech and language, which will continue to accelerate progress towards a clinically viable speech neuroprosthesis.

An accurate and rapidly calibrating speech neuroprosthesis

Brain-computer interfaces can enable rapid, intuitive communication for people with paralysis by transforming the cortical activity associated with attempted speech into text on a computer screen. Despite recent advances, communication with brain-computer interfaces has been restricted by extensive training data requirements and inaccurate word output. A man in his 40's with ALS with tetraparesis and severe dysarthria (ALSFRS-R = 23) was enrolled into the BrainGate2 clinical trial. He underwent surgical implantation of four microelectrode arrays into his left precentral gyrus, which recorded neural activity from 256 intracortical electrodes. We report a speech neuroprosthesis that decoded his neural activity as he attempted to speak in both prompted and unstructured conversational settings. Decoded words were displayed on a screen, then vocalized using text-to-speech software designed to sound like his pre-ALS voice. On the first day of system use, following 30 minutes of attempted speech training data, the neuroprosthesis achieved 99.6% accuracy with a 50-word vocabulary. On the second day, the size of the possible output vocabulary increased to 125,000 words, and, after 1.4 additional hours of training data, the neuroprosthesis achieved 90.2% accuracy. With further training data, the neuroprosthesis sustained 97.5% accuracy beyond eight months after surgical implantation. The participant has used the neuroprosthesis to communicate in self-paced conversations for over 248 hours. In an individual with ALS and severe dysarthria, an intracortical speech neuroprosthesis reached a level of performance suitable to restore naturalistic communication after a brief training period.

Decoding the Spike-Band Subthreshold Motor Cortical Activity

Intracortical Brain-Computer Interfaces (iBCI) use single-unit activity (SUA), multiunit activity (MUA) and local field potentials (LFP) to control neuroprosthetic devices. SUA and MUA are usually extracted from the bandpassed recording through amplitude thresholding, while subthreshold data are ignored. Here, we show that subthreshold data can actually be decoded to determine behavioral variables with test set accuracy of up to 100%. Although the utility of SUA, MUA and LFP for decoding behavioral variables has been explored previously, this study investigates the utility of spike-band subthreshold activity exclusively. We provide evidence suggesting that this activity can be used to keep decoding performance at acceptable levels even when SUA quality is reduced over time. To the best of our knowledge, the signals that we derive from the subthreshold activity may be the weakest neural signals that have ever been extracted from extracellular neural recordings, while still being decodable with test set accuracy of up to 100%. These results are relevant for the development of fully data-driven and automated methods for amplitude thresholding spike-band extracellular neural recordings in iBCIs containing thousands of electrodes.

Recommendations for promoting user agency in the design of speech neuroprostheses

Brain-computer interfaces (BCI) that directly decode speech from brain activity aim to restore communication in people with paralysis who cannot speak. Despite recent advances, neural inference of speech remains imperfect, limiting the ability for speech BCIs to enable experiences such as fluent conversation that promote agency - that is, the ability for users to author and transmit messages enacting their intentions. Here, we make recommendations for promoting agency based on existing and emerging strategies in neural engineering. The focus is on achieving fast, accurate, and reliable performance while ensuring volitional control over when a decoder is engaged, what exactly is decoded, and how messages are expressed. Additionally, alongside neuroscientific progress within controlled experimental settings, we argue that a parallel line of research must consider how to translate experimental successes into real-world environments. While such research will ultimately require input from prospective users, here we identify and describe design choices inspired by human-factors work conducted in existing fields of assistive technology, which address practical issues likely to emerge in future real-world speech BCI applications.

Towards clinical application of implantable brain-computer interfaces for people with late-stage ALS: medical and ethical considerations

Individuals with amyotrophic lateral sclerosis (ALS) frequently develop speech and communication problems in the course of their disease. Currently available augmentative and alternative communication technologies do not present a solution for many people with advanced ALS, because these devices depend on residual and reliable motor activity. Brain-computer interfaces (BCIs) use neural signals for computer control and may allow people with late-stage ALS to communicate even when conventional technology falls short. Recent years have witnessed fast progression in the development and validation of implanted BCIs, which place neural signal recording electrodes in or on the cortex. Eventual widespread clinical application of implanted BCIs as an assistive communication technology for people with ALS will have significant consequences for their daily life, as well as for the clinical management of the disease, among others because of the potential interaction between the BCI and other procedures people with ALS undergo, such as tracheostomy. This article aims to facilitate responsible real-world implementation of implanted BCIs. We review the state of the art of research on implanted BCIs for communication, as well as the medical and ethical implications of the clinical application of this technology. We conclude that the contribution of all BCI stakeholders, including clinicians of the various ALS-related disciplines, will be needed to develop procedures for, and shape the process of, the responsible clinical application of implanted BCIs.

Increasing Robustness of Brain-Computer Interfaces Through Automatic Detection and Removal of Corrupted Input Signals

For brain–computer interfaces (BCIs) to be viable for long-term daily usage, they must be able to quickly identify and adapt to signal disruptions. Furthermore, the detection and mitigation steps need to occur automatically and without the need for user intervention while also being computationally tractable for the low-power hardware that will be used in a deployed BCI system. Here, we focus on disruptions that are likely to occur during chronic use that cause some recording channels to fail but leave the remaining channels unaffected. In these cases, the algorithm that translates recorded neural activity into actions, the neural decoder, should seamlessly identify and adjust to the altered neural signals with minimal inconvenience to the user. First, we introduce an adapted statistical process control (SPC) method that automatically identifies disrupted channels so that both decoding algorithms can be adjusted, and technicians can be alerted. Next, after identifying corrupted channels, we demonstrate the automated and rapid removal of channels from a neural network decoder using a masking approach that does not change the decoding architecture, making it amenable for transfer learning. Finally, using transfer and unsupervised learning techniques, we update the model weights to adjust for the corrupted channels without requiring the user to collect additional calibration data. We demonstrate with both real and simulated neural data that our approach can maintain high-performance while simultaneously minimizing computation time and data storage requirements. This framework is invisible to the user but can dramatically increase BCI robustness and usability.

Dimensionality Reduction of Local Field Potential Features with Convolution Neural Network in Neural Decoding: A Pilot Study

Local field potentials (LFPs) have better long-term stability compared with spikes in brain-machine interfaces (BMIs). Many studies have shown promising results of LFP decoding, but the high-dimensional feature of LFP still hurdle the development of the BMIs to low-cost. In this paper, we proposed a framework of a 1D convolution neural network (CNN) to reduce the dimensionality of the LFP features. For evaluating the performance of this architecture, the reduced LFP features were decoded to cursor position (Center-out task) by a Kalman filter. The Principal components analysis (PCA) was also performed as a comparison. The results showed that the CNN model could reduce the dimensionality of LFP features to a smaller size without significant performance loss. The decoding result based on the CNN features outperformed that based on the PCA features. Moreover, the reduced features by CNN also showed robustness across different sessions. These results demonstrated that the LFP features reduced by the CNN model achieved low cost without sacrificing high-performance and robustness, suggesting that this method could be used for portable BMI systems in the future.

Implantable brain-computer interface for neuroprosthetic-enabled volitional hand grasp restoration in spinal cord injury

Loss of hand function after cervical spinal cord injury severely impairs functional independence. We describe a method for restoring volitional control of hand grasp in one 21-year-old male subject with complete cervical quadriplegia (C5 American Spinal Injury Association Impairment Scale A) using a portable fully implanted brain-computer interface within the home environment. The brain-computer interface consists of subdural surface electrodes placed over the dominant-hand motor cortex and connects to a transmitter implanted subcutaneously below the clavicle, which allows continuous reading of the electrocorticographic activity. Movement-intent was used to trigger functional electrical stimulation of the dominant hand during an initial 29-weeks laboratory study and subsequently via a mechanical hand orthosis during in-home use. Movement-intent information could be decoded consistently throughout the 29-weeks in-laboratory study with a mean accuracy of 89.0% (range 78-93.3%). Improvements were observed in both the speed and accuracy of various upper extremity tasks, including lifting small objects and transferring objects to specific targets. At-home decoding accuracy during open-loop trials reached an accuracy of 91.3% (range 80-98.95%) and an accuracy of 88.3% (range 77.6-95.5%) during closed-loop trials. Importantly, the temporal stability of both the functional outcomes and decoder metrics were not explored in this study. A fully implanted brain-computer interface can be safely used to reliably decode movement-intent from motor cortex, allowing for accurate volitional control of hand grasp.

Spinal Cord Injury-Induced Changes in Encoding and Decoding of Bipedal Walking by Motor Cortical Ensembles

Recent studies have shown that motor recovery following spinal cord injury (SCI) is task-specific. However, most consequential conclusions about locomotor functional recovery from SCI have been derived from quadrupedal locomotion paradigms. In this study, two monkeys were trained to perform a bipedal walking task, mimicking human walking, before and after T8 spinal cord hemisection. Importantly, there is no pharmacological therapy with nerve growth factor for monkeys after SCI; thus, in this study, the changes that occurred in the brain were spontaneous. The impairment of locomotion on the ipsilateral side was more severe than that on the contralateral side. We used information theory to analyze single-cell activity from the left primary motor cortex (M1), and results show that neuronal populations in the unilateral primary motor cortex gradually conveyed more information about the bilateral hindlimb muscle activities during the training of bipedal walking after SCI. We further demonstrated that, after SCI, progressively expanded information from the neuronal population reconstructed more accurate control of muscle activity. These results suggest that, after SCI, the unilateral primary motor cortex could gradually regain control of bilateral coordination and motor recovery and in turn enhance the performance of brain–machine interfaces.

Generalizable cursor click decoding using grasp-related neural transients

Objective.Intracortical brain-computer interfaces (iBCI) have the potential to restore independence for individuals with significant motor or communication impairments. One of the most realistic avenues for clinical translation of iBCI technology is enabling control of a computer cursor-i.e. movement-related neural activity is interpreted (decoded) and used to drive cursor function. Here we aim to improve cursor click decoding to allow for both point-and-click and click-and-drag control.Approach.Using chronic microelectrode arrays implanted in the motor cortex of two participants with tetraplegia, we identified prominent neural responses related to attempted hand grasp. We then developed a new approach for decoding cursor click (hand grasp) based on the most salient responses.Main results.We found that the population-wide response contained three dominant components related to hand grasp: an onset transient response, a sustained response, and an offset transient response. The transient responses were larger in magnitude-and thus more reliably detected-than the sustained response, and a click decoder based on these transients outperformed the standard approach of binary state classification.Significance.A transient-based approach for identifying hand grasp can provide a high degree of cursor click control for both point-and-click and click-and-drag applications. This generalized click functionality is an important step toward high-performance cursor control and eventual clinical translation of iBCI technology.

Long-term intracortical microelectrode array performance in a human: a 5 year retrospective analysis

Objective. Brain-computer interfaces (BCIs) that record neural activity using intracortical microelectrode arrays (MEAs) have shown promise for mitigating disability associated with neurological injuries and disorders. While the chronic performance and failure modes of MEAs have been well studied and systematically described in non-human primates, there is far less reported about long-term MEA performance in humans. Our group has collected one of the largest neural recording datasets from a Utah MEA in a human subject, spanning over 5 years (2014-2019). Here we present both long-term signal quality and BCI performance as well as highlight several acute signal disruption events observed during the clinical study.Approach. Long-term Utah array performance was evaluated by analyzing neural signal metric trends and decoding accuracy for tasks regularly performed across 448 clinical recording sessions. For acute signal disruptions, we identify or hypothesize the root cause of the disruption, show how the disruption manifests in the collected data, and discuss potential identification and mitigation strategies for the disruption.Main results. Neural signal quality metrics deteriorated rapidly within the first year, followed by a slower decline through the remainder of the study. Nevertheless, BCI performance remained high 5 years after implantation, which is encouraging for the translational potential of this technology as an assistive device. We also present examples of unanticipated signal disruptions during chronic MEA use, which are critical to detect as BCI technology progresses toward home usage.Significance. Our work fills a gap in knowledge around long-term MEA performance in humans, providing longevity and efficacy data points to help characterize the performance of implantable neural sensors in a human population. The trial was registered on ClinicalTrials.gov (Identifier NCT01997125) and conformed to institutional requirements for the conduct of human subjects research.

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.

Classifying Intracortical Brain-Machine Interface Signal Disruptions Based on System Performance and Applicable Compensatory Strategies: A Review

Brain-machine interfaces (BMIs) record and translate neural activity into a control signal for assistive or other devices. Intracortical microelectrode arrays (MEAs) enable high degree-of-freedom BMI control for complex tasks by providing fine-resolution neural recording. However, chronically implanted MEAs are subject to a dynamic in vivo environment where transient or systematic disruptions can interfere with neural recording and degrade BMI performance. Typically, neural implant failure modes have been categorized as biological, material, or mechanical. While this categorization provides insight into a disruption's causal etiology, it is less helpful for understanding degree of impact on BMI function or possible strategies for compensation. Therefore, we propose a complementary classification framework for intracortical recording disruptions that is based on duration of impact on BMI performance and requirement for and responsiveness to interventions: (1) Transient disruptions interfere with recordings on the time scale of minutes to hours and can resolve spontaneously; (2) Reversible disruptions cause persistent interference in recordings but the root cause can be remedied by an appropriate intervention; (3) Irreversible compensable disruptions cause persistent or progressive decline in signal quality, but their effects on BMI performance can be mitigated algorithmically; and (4) Irreversible non-compensable disruptions cause permanent signal loss that is not amenable to remediation or compensation. This conceptualization of intracortical BMI disruption types is useful for highlighting specific areas for potential hardware improvements and also identifying opportunities for algorithmic interventions. We review recording disruptions that have been reported for MEAs and demonstrate how biological, material, and mechanical mechanisms of disruption can be further categorized according to their impact on signal characteristics. Then we discuss potential compensatory protocols for each of the proposed disruption classes. Specifically, transient disruptions may be minimized by using robust neural decoder features, data augmentation methods, adaptive machine learning models, and specialized signal referencing techniques. Statistical Process Control methods can identify reparable disruptions for rapid intervention. In-vivo diagnostics such as impedance spectroscopy can inform neural feature selection and decoding models to compensate for irreversible disruptions. Additional compensatory strategies for irreversible disruptions include information salvage techniques, data augmentation during decoder training, and adaptive decoding methods to down-weight damaged channels.