Direct brain control of overground walking in those with paraplegia due to spinal cord injury (SCI) has not been achieved. Invasive brain-computer interfaces (BCIs) may provide a permanent solution to this problem by directly linking the brain to lower extremity prostheses. To justify the pursuit of such invasive systems, the feasibility of BCI controlled overground walking should first be established in a noninvasive manner. To accomplish this goal, we developed an electroencephalogram (EEG)-based BCI to control a functional electrical stimulation (FES) system for overground walking and assessed its performance in an individual with paraplegia due to SCI.
BACKGROUND: We developed an electroencephalogram-based brain computer interface system to modulate functional electrical stimulation (FES) to the affected tibialis anterior muscle in a stroke patient. The intensity of FES current increased in a stepwise manner when the event-related desynchronization (ERD) reflecting motor intent was continuously detected from the primary cortical motor area. METHODS: We tested the feasibility of the ERD-modulated FES system in comparison with FES without ERD modulation. The stroke patient who presented with severe hemiparesis attempted to perform dorsiflexion of the paralyzed ankle during which FES was applied either with or without ERD modulation. RESULTS: After 20 minutes of training, the range of movement at the ankle joint and the electromyography amplitude of the affected tibialis anterior muscle were significantly increased following the ERD-modulated FES compared with the FES alone. CONCLUSIONS: The proposed rehabilitation technique using ERD-modulated FES for stroke patients was feasible. The system holds potentials to improve the limb function and to benefit stroke patients.
Brain-machine interfaces (BMI) usually decode movement parameters from cortical activity to control neuroprostheses. This requires subjects to learn to modulate their brain activity to convey all necessary information, thus imposing natural limits on the complexity of tasks that can be performed. Here we demonstrate an alternative and complementary BMI paradigm that overcomes that limitation by decoding cognitive brain signals associated with monitoring processes relevant for achieving goals. In our approach the neuroprosthesis executes actions that the subject evaluates as erroneous or correct, and exploits the brain correlates of this assessment to learn suitable motor behaviours. Results show that, after a short user’s training period, this teaching BMI paradigm operated three different neuroprostheses and generalized across several targets. Our results further support that these error-related signals reflect a task-independent monitoring mechanism in the brain, making this teaching paradigm scalable. We anticipate this BMI approach to become a key component of any neuroprosthesis that mimics natural motor control as it enables continuous adaptation in the absence of explicit information about goals. Furthermore, our paradigm can seamlessly incorporate other cognitive signals and conventional neuroprosthetic approaches, invasive or non-invasive, to enlarge the range and complexity of tasks that can be accomplished.
Localization in Reverberation with Cochlear Implants : Predicting Performance from Basic Psychophysical Measures
- Journal of the Association for Research in Otolaryngology : JARO
- Published about 8 years ago
Users of bilateral cochlear implants (CIs) experience difficulties localizing sounds in reverberant rooms, even in rooms where normal-hearing listeners would hardly notice the reverberation. We measured the localization ability of seven bilateral CI users listening with their own devices in anechoic space and in a simulated reverberant room. To determine factors affecting performance in reverberant space we measured the sensitivity to interaural time differences (ITDs), interaural level differences (ILDs), and forward masking in the same participants using direct computer control of the electric stimulation in their CIs. Localization performance, quantified by the coefficient of determination r (2) and the root mean squared error, was significantly worse in the reverberant room than in anechoic conditions. Localization performance in the anechoic room, expressed as r (2), was best predicted by subject’s sensitivity to ILDs. However, the decrease in localization performance caused by reverberation was better predicted by the sensitivity to envelope ITDs measured on single electrode pairs, with a correlation coefficient of 0.92. The CI users who were highly sensitive to envelope ITDs also better maintained their localization ability in reverberant space. Results in the forward masking task added only marginally to the predictions of localization performance in both environments. The results indicate that envelope ITDs provided by CI processors support localization in reverberant space. Thus, methods that improve perceptual access to envelope ITDs could help improve localization with bilateral CIs in everyday listening situations.
Abdominal functional electrical stimulation (abdominal FES) is the application of a train of electrical pulses to the abdominal muscles, causing them to contract. Abdominal FES has been used as a neuroprosthesis to acutely augment respiratory function and as a rehabilitation tool to achieve a chronic increase in respiratory function after abdominal FES training, primarily focusing on patients with spinal cord injury (SCI). This study aimed to review the evidence surrounding the use of abdominal FES to improve respiratory function in both an acute and chronic manner after SCI.
The mechanical mismatch between soft neural tissues and stiff neural implants hinders the long-term performance of implantable neuroprostheses. Here, we designed and fabricated soft neural implants with the shape and elasticity of dura mater, the protective membrane of the brain and spinal cord. The electronic dura mater, which we call e-dura, embeds interconnects, electrodes, and chemotrodes that sustain millions of mechanical stretch cycles, electrical stimulation pulses, and chemical injections. These integrated modalities enable multiple neuroprosthetic applications. The soft implants extracted cortical states in freely behaving animals for brain-machine interface and delivered electrochemical spinal neuromodulation that restored locomotion after paralyzing spinal cord injury.
Neuroprosthetics that combine a brain computer interface (BCI) with functional electrical stimulation (FES) can restore voluntary control of a patients' own paralyzed limbs. To date, human studies have demonstrated an “all-or-none” type of control for a fixed number of pre-determined states, like hand-open and hand-closed. To be practical for everyday use, a BCI-FES system should enable smooth control of limb movements through a continuum of states and generate situationally appropriate, graded muscle contractions. Crucially, this functionality will allow users of BCI-FES neuroprosthetics to manipulate objects of different sizes and weights without dropping or crushing them. In this study, we present the first evidence that using a BCI-FES system, a human with tetraplegia can regain volitional, graded control of muscle contraction in his paralyzed limb. In addition, we show the critical ability of the system to generalize beyond training states and accurately generate wrist flexion states that are intermediate to training levels. These innovations provide the groundwork for enabling enhanced and more natural fine motor control of paralyzed limbs by BCI-FES neuroprosthetics.
Target identification and contact selection are known contributors to variability in efficacy across different clinical indications of deep brain stimulation surgery. A retrospective analysis of responders to subcallosal cingulate deep brain stimulation (SCC DBS) for depression demonstrated the common impact of the electrical stimulation on a stereotypic connectome of converging white matter bundles (forceps minor, uncinate fasciculus, cingulum and fronto-striatal fibers). To test the utility of a prospective connectomic approach for SCC DBS surgery, this pilot study used the four-bundle tractography ‘connectome blueprint’ to plan surgical targeting in 11 participants with treatment-resistant depression. Before surgery, targets were selected individually using deterministic tractography. Selection of contacts for chronic stimulation was made by matching the post-operative probabilistic tractography map to the pre-surgical deterministic tractography map for each subject. Intraoperative behavioral responses were used as a secondary verification of location. A probabilistic tract map of all participants demonstrated inclusion of the four bundles as intended, matching the connectome blueprint previously defined. Eight of 11 patients (72.7%) were responders and 5 were remitters after 6 months of open-label stimulation. At one year, 9 of 11 patients (81.8%) were responders, with 6 of them in remission. These results support the utility of a group probabilistic tractography map as a connectome blueprint for individualized, patient-specific, deterministic tractography targeting, confirming retrospective findings previously published. This new method represents a connectomic approach to guide future SCC DBS studies.Molecular Psychiatry advance online publication, 11 April 2017; doi:10.1038/mp.2017.59.
Neuroprosthetic brain-computer interfaces function via an algorithm which decodes neural activity of the user into movements of an end effector, such as a cursor or robotic arm. In practice, the decoder is often learned by updating its parameters while the user performs a task. When the user’s intention is not directly observable, recent methods have demonstrated value in training the decoder against a surrogate for the user’s intended movement. Here we show that training a decoder in this way is a novel variant of an imitation learning problem, where an oracle or expert is employed for supervised training in lieu of direct observations, which are not available. Specifically, we describe how a generic imitation learning meta-algorithm, dataset aggregation (DAgger), can be adapted to train a generic brain-computer interface. By deriving existing learning algorithms for brain-computer interfaces in this framework, we provide a novel analysis of regret (an important metric of learning efficacy) for brain-computer interfaces. This analysis allows us to characterize the space of algorithmic variants and bounds on their regret rates. Existing approaches for decoder learning have been performed in the cursor control setting, but the available design principles for these decoders are such that it has been impossible to scale them to naturalistic settings. Leveraging our findings, we then offer an algorithm that combines imitation learning with optimal control, which should allow for training of arbitrary effectors for which optimal control can generate goal-oriented control. We demonstrate this novel and general BCI algorithm with simulated neuroprosthetic control of a 26 degree-of-freedom model of an arm, a sophisticated and realistic end effector.
Studies using vocoders as acoustic simulators of cochlear implants have generally focused on simulation of speech understanding, gender recognition, or music appreciation. The aim of the present experiment was to study the auditory sensation perceived by cochlear implant (CI) recipients with steady electrical stimulation on the most-apical electrode.