Lee E. Miller

Restoration of grasp following paralysis through brain-controlled stimulation of muscles

Patients with spinal cord injury lack the connections between brain and spinal cord circuits that are essential for voluntary movement. Clinical systems that achieve muscle contraction through functional electrical stimulation (FES) have proven to be effective in allowing patients with tetraplegia to regain control of hand movements and to achieve a greater measure of independence in daily activities. In existing clinical systems, the patient uses residual proximal limb movements to trigger pre-programmed stimulation that causes the paralysed muscles to contract, allowing use of one or two basic grasps. Instead, we have developed an FES system in primates that is controlled by recordings made from microelectrodes permanently implanted in the brain. We simulated some of the effects of the paralysis caused by C5 or C6 spinal cord injury by injecting rhesus monkeys with a local anaesthetic to block the median and ulnar nerves at the elbow. Then, using recordings from approximately 100 neurons in the motor cortex, we predicted the intended activity of several of the paralysed muscles, and used these predictions to control the intensity of stimulation of the same muscles. This process essentially bypassed the spinal cord, restoring to the monkeys voluntary control of their paralysed muscles. This achievement is a major advance towards similar restoration of hand function in human patients through brain-controlled FES. We anticipate that in human patients, this neuroprosthesis would allow much more flexible and dexterous use of the hand than is possible with existing FES systems.

Primary motor cortical neurons encode functional muscle synergies

Many different kinematic and kinetic signals have been proposed as possible variables under the control of the primary motor cortex. Despite the presence of direct projections to motor neurons, muscle activation has received less attention as a controlled variable. Furthermore, although it is well known that descending fibers project to multiple motor pools, an objective, quantitative study of the relation between neuronal modulation and the activity of groups of muscles has not previously been reported. We have recorded the discharge of 310 neurons located in the primary motor cortex of two monkeys, along with the activity of a variety of arm and hand muscles. Data were recorded while the monkey reached to and pressed a series of illuminated buttons. The similarity of a given neuron’s discharge with respect to each muscle was determined by calculating the linear cross-correlation between its discharge rate and each rectified, filtered electromyogram. A “functional linkage vector” was then constructed, which expressed the similarity of that neuron’s discharge to the entire set of muscles. We discovered discrete groups of functional linkage vectors within the high order muscle space for both monkeys which corresponded to functional properties of the neurons measured by other methods. Several of these groups appeared to represent a functional synergy of muscles, such as those required to extend the limb, press a button, or open the hand in preparation for the press. When the dimensionality of this space was reduced by a principal components analysis, the originally identified clusters of neurons remained well separated. These results are consistent with the hypothesis that the discharge of individual neurons in the primary motor cortex encodes the activity of a relatively small number of functionally relevant groups of muscles. It will be important to determine whether these results will also apply to more complex behavior, and to what extent these functional muscle synergy representations remain fixed across behaviors.

Restoring sensorimotor function through intracortical interfaces: progress and looming challenges

The loss of a limb or paralysis resulting from spinal cord injury has devastating consequences on quality of life. One approach to restoring lost sensory and motor abilities in amputees and patients with tetraplegia is to supply them with implants that provide a direct interface with the CNS. Such brain–machine interfaces might enable a patient to exert voluntary control over a prosthetic or robotic limb or over the electrically induced contractions of paralysed muscles. A parallel interface could convey sensory information about the consequences of these movements back to the patient. Recent developments in the algorithms that decode motor intention from neuronal activity and in approaches to convey sensory feedback by electrically stimulating neurons, using biomimetic and adaptation-based approaches, have shown the promise of invasive interfaces with sensorimotor cortices, although substantial challenges remain.

Toward the Restoration of Hand Use to a Paralyzed Monkey: Brain-Controlled Functional Electrical Stimulation of Forearm Muscles

Loss of hand use is considered by many spinal cord injury survivors to be the most devastating consequence of their injury. Functional electrical stimulation (FES) of forearm and hand muscles has been used to provide basic, voluntary hand grasp to hundreds of human patients. Current approaches typically grade pre-programmed patterns of muscle activation using simple control signals, such as those derived from residual movement or muscle activity. However, the use of such fixed stimulation patterns limits hand function to the few tasks programmed into the controller. In contrast, we are developing a system that uses neural signals recorded from a multi-electrode array implanted in the motor cortex; this system has the potential to provide independent control of multiple muscles over a broad range of functional tasks. Two monkeys were able to use this cortically controlled FES system to control the contraction of four forearm muscles despite temporary limb paralysis. The amount of wrist force the monkeys were able to produce in a one-dimensional force tracking task was significantly increased. Furthermore, the monkeys were able to control the magnitude and time course of the force with sufficient accuracy to track visually displayed force targets at speeds reduced by only one-third to one-half of normal. Although these results were achieved by controlling only four muscles, there is no fundamental reason why the same methods could not be scaled up to control a larger number of muscles. We believe these results provide an important proof of concept that brain-controlled FES prostheses could ultimately be of great benefit to paralyzed patients with injuries in the mid-cervical spinal cord.

Correlation of primate red nucleus discharge with muscle activity during free-form arm movements.

1. We recorded from 239 neurons located in the magnocellular division of the red nucleus of four alert macaque monkeys. At the same time, we recorded electromyographic (EMG) signals from as many as twenty electrodes chronically implanted on muscles of the shoulder, arm, forearm and hand. We recorded EMG signals for periods ranging from several months to a year. 2. The monkeys were trained to perform three free‐form food retrieval tasks, each of which activated all of the recorded muscles and most of the neurons. The ‘prehension’ task required simply that the monkey grasp a piece of food from a fixed point in space. The ‘barrier’ task required the monkey to reach around a small barrier to obtain the food, and the ‘Kluver’ task required that food be removed from small holes. During the prehension task, we found approximately equal numbers of neurons that were strongly active while the hand was being moved toward the target (70% of units), and while the food was being grasped (60%). Relatively few units were active as the hand was returned to the mouth (15%). 3. Data files of 1‐2 min duration were collected while the monkey performed a single behavioural task. Whenever possible, we recorded files for all three tasks from each neuron. For each file we calculated long time‐span analog cross‐correlations (+/‐ 1.28 s) between instantaneous neuronal firing rate and each of the full‐wave rectified, low‐pass filtered EMG signals. We used the peak correlation and the time of the peak as two summary measures of the functional relation between modulation of neuronal activity and EMG. 4. The magnitude of the strongest correlations was between 0.4 and 0.5 (normalized to a perfect correlation of +/‐ 1.0). Distal muscles were the most frequently correlated, and extensors were more frequently correlated than flexors. For all monkeys, the lags for well correlated muscles were distributed broadly about a uni‐modal value near 0 ms. Eighty five per cent of the correlations larger than or equal to 0.25 had peaks between ‐150 and 200 ms. 5. The activity of each neuron was represented in a muscle co‐ordinate system by an n‐dimensional ‘functional linkage vector’, each element of which was the peak correlation with one of n muscles. The vector for any given neuron points in a particular direction in muscle space, depending on the similarity between the activity of the neuron and the activity of each muscle.(ABSTRACT TRUNCATED AT 400 WORDS)

The use of overlapping submovements in the control of rapid hand movements

Rapid targeted movements are subject to special control considerations, since there may be inadequate time available for either visual or somatosensory feedback to be effective. In our experiments, subjects rapidly rotated a knob to align a pointer to one of several targets. We recognized three different types of movement segments: the primary movement, and two types of submovement, which frequently followed. The submovements were initiated either before or after the end of the primary movement. The former, or “overlapping” type of submovement altered the kinematics of the overall movement and was consequently difficult to detect. We used a direct, objective test of movement regularity to detect overlapping submovements, namely, examining the number of jerk and snap zero crossings during the second half of a movement. Any overlapping submovements were parsed from the overall movement by subtracting the velocity profile of the primary movement. The velocity profiles of the extracted submovements had near-symmetric bell shapes, similar to the shapes of both pure primary movements and nonoverlapping submovements. This suggests that the same neural control mechanisms may be responsible for producing all three types of movement segments. Overlapping submovements corrected for errors in the amplitude of the primary movement. Furthermore, they may account for the previously observed, speed-dependent asymmetry of the velocity profile. We used a nonlinear model of the musculoskeletal system to explain most of the kinematic features of these rapid hand movements, including how discrete submovements are superimposed on a primary movement. Finally, we present a plausible scheme for how the central nervous system may generate the commands to control these rapid hand movements.

Accurate decoding of reaching movements from field potentials in the absence of spikes

The recent explosion of interest in brain-machine interfaces (BMIs) has spurred research into choosing the optimal input signal source for a desired application. The signals with highest bandwidth--single neuron action potentials or spikes--typically are difficult to record for more than a few years after implantation of intracortical electrodes. Fortunately, field potentials recorded within the cortex (local field potentials, LFPs), at its surface (electrocorticograms, ECoG) and at the dural surface (epidural, EFPs) have also been shown to contain significant information about movement. However, the relative performance of these signals has not yet been directly compared. Furthermore, while it is widely postulated, it has not yet been demonstrated that these field potential signals are more durable than spike recordings. The aim of this study was to address both of these questions. We assessed the offline decoding performance of EFPs, LFPs and spikes, recorded sequentially, in primary motor cortex (M1) in terms of their ability to decode the target of reaching movements, as well as the endpoint trajectory. We also examined the decoding performance of LFPs on electrodes that are not recording spikes, compared with the performance when they did record spikes. Spikes were still present on some of the other electrodes throughout this study. We showed that LFPs performed nearly as well as spikes in decoding velocity, and slightly worse in decoding position and in target classification. EFP performance was slightly inferior to that reported for ECoG in humans. We also provided evidence demonstrating that movement-related information in the LFP remains high regardless of the ability to record spikes concurrently on the same electrodes. This is the first study to provide evidence that LFPs retain information about movement in the absence of spikes on the same electrodes. These results suggest that LFPs may indeed remain informative after spike recordings are lost, thereby providing a robust, accurate signal source for BMIs.

Optimal spacing of surface electrode arrays for brain–machine interface applications

Brain-machine interfaces (BMIs) use signals recorded directly from the brain to control an external device, such as a computer cursor or a prosthetic limb. These control signals have been recorded from different levels of the brain, from field potentials at the scalp or cortical surface to single neuron action potentials. At present, the more invasive recordings have better signal quality, but also lower stability over time. Recently, subdural field potentials have been proposed as a stable, good quality source of control signals, with the potential for higher spatial and temporal bandwidth than EEG. Here we used finite element modeling in rats and humans and spatial spectral analysis in rats to compare the spatial resolution of signals recorded epidurally (outside the dura), with those recorded from subdural and scalp locations. Resolution of epidural and subdural signals was very similar in rats and somewhat less so in human models. Both were substantially better than signals recorded at the scalp. Resolution of epidural and subdural signals in humans was much more similar when the cerebrospinal fluid layer thickness was reduced. This suggests that the less invasive epidural recordings may yield signals of similar quality to subdural recordings, and hence may be more attractive as a source of control signals for BMIs.

The relative activation of muscles during isometric contractions and low-velocity movements against a load.

Surface electromyographic (EMG) and motor unit activity were measured in human arm muscles during isometric contractions and during movements against an elastic load. The direction of force applied proximal to the wrist and movement direction of the wrist were varied in a horizontal plane. During isometric contractions the direction in which the largest EMG activity was measured corresponded to the direction in which motor units had the smallest recruitment threshold, for each muscle. The same was found for movements against an elastic load. However, this direction was different for isometric contractions and for movements. Because the magnitude and sign of these changes varied for different muscles, this resulted in a different relative activation of muscles for the two conditions. The amplitude of the surface EMG during contractions against an elastic load was generally significantly larger than that for isometric contractions against the same load. For m. brachioradialis isometric conditions yielded occasionally increased EMG activity. The change in EMG activity could be attributed completely to changes in motor unit recruitment thresholds leading to proportionate changes in the number of recruited motor units. However, the initial firing rate of motor units at recruitment was the same under both conditions and, therefore, did not contribute to changes in amplitude of surface EMG activity.

Electrical Stimulation of the Proprioceptive Cortex (Area 3a) Used to Instruct a Behaving Monkey

A growing number of brain-machine interfaces have now been developed that allow movements of an external device to be controlled using recordings from the brain. This work has been undertaken with a number of different animal models, as well as several human patients with quadriplegia. The resulting movements, whether of computer cursors or robotic limbs, remain quite slow and unstable compared to normal limb movements. It is an open question, how much of this instability is the result of the limited forward control path, and how much has to do with the total lack of normal proprioceptive feedback. We have begun preliminary studies of the effectiveness of electrical stimulation in the proprioceptive area of the primary somatosensory cortex (area 3a) as a potential means to deliver an artificial sense of proprioception to a monkey. We have demonstrated that it is possible for the monkey to detect brief stimulus trains at relatively low current levels, and to discriminate between trains of different frequencies. These observations need to be expanded to include more complex, time-varying waveforms that could potentially convey information about the state of the limb.