Andrew Fuglevand - US grants

DESCRIPTION (provided by applicant): The ability to move the fingers relatively independently of one another enables humans and non-human primates to manipulate diverse objects in the environment and to execute an immense variety of movements and gestures. Lesions of the motor cortex, such as those associated with stroke, give rise to persistent deficits in the ability to move the fingers individually even though the capacity to control other limb segments recovers. Such severe impairment in finger dexterity following stroke might arise because of the unusually complex coordination required among multiple muscles to produce even simple finger movements. However, little is known about how the central nervous system orchestrates ensembles of muscles to 3roduce movements of the fingers and hand. The broad goal of this project, therefore, is to identify the neural mechanisms whereby the activities of multiple muscles are coordinated in the elaboration of finger movements. Specifically, cross-correlation analysis of the firing times of motor units located in different hand muscles of human subjects will be used to estimate the extent to which coordinated activity among muscles is due to divergence of descending pathways providing common synpatic input across sets of motor nuclei. The specific aims are designed to identify how hand muscles are assembled into functional groups for four widely different forms of synergistic activity that underlie: 1) the coordination of compartments of multi-tendoned muscles that insert upon multiple digits, 2) the coordination of separate muscles that insert upon the same digit, 3) the coordination of separate muscles that insert upon different digits yet are used habitually together for specific functions, and 4) the coordination between prime movers of digits and muscles whose activity provides joint stabilization necessary for effective digit movements. These studies will expand our knowledge of the mechanisms by which the brain controls voluntary movements and may provide important insights into the causes of motor dysfunction associated with stroke.

[unreadable] DESCRIPTION (provided by applicant): Functional electrical stimulation involves artificial activation of paralyzed muscles with implanted electrodes and has been used successfully to improve the ability of quadriplegics to perform movements important for daily activities. The range of motor behaviors that can be generated by functional electrical stimulation, however, is limited to a relatively small set of preprogrammed movements such as hand grasp and release. A broader range of movements has not been implemented because of the substantial challenge associated with identifying the patterns of muscle stimulation needed to elicit specified movements. We plan to use a probabilistic algorithm to predict the patterns of muscle activity associated with a wide range of upper limb movements based on hand trajectory information. The predicted patterns of muscle activity will then be transformed into amplitude-modulated trains of pulses and used to drive muscle stimulators in order to evoke movements in temporarily paralyzed animals. The evoked movements are quantitatively compared to the desired movements to evaluate the overall effectiveness of this approach. Ultimately, this probabilistic method could serve as the requisite interface between brain-derived trajectory information and existing functional electrical stimulation systems to realize a self-contained and self-controlled upper limb neuroprosthetic system. Such an integrated and flexible system would greatly increase movement capability, and independence, in paralyzed individuals. PUBLIC HEALTH RELEVANCE: The goal of this project is to develop a new method to artificially activate and control paralyzed muscles with electrodes implanted in muscles. This effort will contribute to the restoration of voluntary limb movements in individuals paralyzed because of spinal cord injury or stroke. [unreadable] [unreadable] [unreadable]

A typical neuron continuously processes synaptic inputs from thousands of other neurons and converts these inputs into reliable frequency-coded messages that are relayed to other neurons and effectors. Establishing how individual neurons integrate such complex input is essential for understanding how the brain processes information, yet surprisingly little is known about how this fundamental process operates. This is partly due to the difficulty of experimentally controlling the intensity of synaptic input to a neuron. Experimentalists instead rely on artificial excitation by injecting current pulses into neurons through microelectrodes. Such current-injection methods may not validly represent the actual process of synaptic integration. Furthermore, transformation of synaptic input into a firing rate response involves complex interactions among passive and active membrane properties. We propose to characterize basic features of synaptic integration by recording the firing-rate responses of human motor neurons to relatively controlled levels of synaptic input. Motor neurons are particularly attractive for these purposes because their activities can be readily recorded in human subjects, the synaptic drive to motor neurons can be easily controlled, and such activity transpires within an intact, un-anesthetized nervous system. We propose three specific aims to address the following questions: 1) what is the nature of the rate-coding response over the entire range of voluntary synaptic drive?, 2) what mechanisms underlie the leveling-off (saturation) of firing rate at low levels in response to increased intensity of synaptic drive?, and 3) can chronic alteration in motor unit activity lead to plastic changes in synaptic integration? Answering these questions will provide new insights into the fundamental process by which neurons transform synaptic input into frequency-coded responses. This information is essential for understanding how the brain operates - a prerequisite for understanding and treating a host of disorders of the nervous system.

DESCRIPTION (provided by applicant): Motor neurons receive synaptic inputs from many other neurons and convert these inputs into frequency-coded messages that are relayed to muscle fibers to cause contraction. It is often assumed that motor neurons generate spikes at rates in proportion to the excitatory synaptic input received. It is now recognized, however, that motor neurons have active processes, such as persistent inward currents (PICs) that may markedly modulate the relationship between synaptic input and firing rate output. PICs represent an intrinsic source of membrane depolarization that may lead to self-sustained firing of motor neurons, i.e., prolonged spiking in the absence of synaptic input. A number of ideas have been forwarded as to the functional significance of PICs, both in terms of the control of normal motor function and as an impaired process contributing to spasticity or amyotrophic lateral sclerosis (ALS). Yet, little is known about the actual physiological conditions under which PICs are activated. Recently, however, a method has been proposed to enable assessment of PIC activation in awake human subjects that involves quantifying an index referred to as ¿F from the activities of pairs of motor units recorded during voluntary contractions. The first specific aim of this project is to rigorously test the validity of the ¿F method based on insights gained from computer modeling. For this aim, we will measure ¿F for pairs of motor units during contractions that vary in rate of rise of force and duration in four muscles whose motor neurons are thought to possess differing capacities for generating PICs. The second specific aim will determine whether the initial high gain in motor unit firing rate observed during voluntary contraction is likely caused by PIC activation. For this aim, we will attempt to prevent activation of PICs altogether by artificially activating strong inhibitory inputs to motor neurons and determine whether this eliminates the initial steep rise in motor unit firing rate. Overall, this project is important because it will provide insight into the physiological conditions that activat PICs. Such information is crucial not only for understanding the fundamental operation of motor neurons but also for identifying the causes of neurological disorders such as spasticity and ALS.

Functional electrical stimulation involves artificial activation of paralyzed muscles with implanted electrodes and has been used successfully to improve the ability of quadriplegics to perform movements important for daily activities. The range of motor behaviors that can be generated by functional electrical stimulation, however, is limited to a relatively small set of preprogrammed movements such as hand grasp and release. A broader range of movements has not been implemented because of the substantial challenge associated with identifying the patterns of muscle stimulation needed to elicit specified movements. To address this limitation, we have developed machine-learning based algorithms that can predict patterns of muscle activity associated with a wide range of complex limb movements. In addition, we have devised a method whereby predicted patterns of muscle activity can then be transformed into stimulus pulse patterns needed to evoke movements in paralyzed limbs. Our goal for this project is to determine whether these approaches, when applied to temporarily paralyzed non-human primates, can be used to produce: 1) a wide range of movements of the hand throughout peri-personal reach space, and 2) configuration of the hand and fingers into a variety of shapes needed to interact with diverse objects in the environment. If successful, this approach would greatly expand the repertoire of motor behaviors available to individuals paralyzed because of spinal cord injury or stroke. Furthermore, this system ultimately might serve as the requisite interface between brain-derived trajectory information and functional electrical stimulation systems needed to realize a self-contained and self- controlled upper limb neuroprosthetic.

ABSTRACT We request a supplement to enable successful continuation and completion of the project Aims as originally outlined in our funded grant proposal (1 R01 NS102259-01A1). This project involves restoration of upper limb movements in temporarily paralyzed non-human primates using machine-learning control of electrical stimulation of 30 muscles. Unanticipated circumstances led to financial challenges that limit our ability to complete the project. If a supplement is provided, we believe that we can successfully complete the Aims of the project. The supplement would provide funds to cover the costs associated with purchasing, training, implanting, and recording from two additional monkeys