Adaptation in Neural Systems
Adaptive processes in neural systems enable biological systems to learn to interpret sensory information and to produce functional movements. At ANS, we are developing therapies and technologies that seek to guide these adaptive processes in order to maximize the recovery of function after trauma or neurological disorders.
R&D projects in this area include:
- neuromotor therapy after spinal cord injury
- therapy for improved function after cerebral palsy
- co-adaptation in engineered and neural systems
Exercise training in Parkinson’s disease: Neural and functional benefits
Key Personnel: Narayanan Krishnamurthi, PhD (PI); ANS, ASU
James Abbas, PhD (ANS/Bioengineering, ASU)
Wayne Willis, PhD (Kinesiology, ASU)
Holly Shill, MD (Sun Health Research Institute)
Kewei Chen, PhD (Banner Good Samaritan Medical Center)
Padma Mahant, MD (Banner Good Samaritan Medical Center)
Johan Samanta, MD (Banner Good Samaritan Medical Center)
Abraham Lieberman, MD (Barrow Neurological Institute)
Sponsor: NIH – National Center for Medical Rehabilitation Research
09/23/08 – 08/31/10, $495,525
For individuals with Parkinson’s disease (PD), symptoms such as tremor, rigidity, poor balance, and difficulties in walking often limit their participation in physical activities. Such reductions in physical activity can accelerate aging-related degradations in motor control and therefore contribute to a spiraling decline in functional capacity that may eventually result in falls, loss of independence, and or respiratory problems. Pharmacological treatment of PD can provide symptomatic relief during the early years, but is often less effective in the advanced stages of the disease and can often produce debilitating side effects. Moreover, there are no treatments currently available to stop or slow the progression of PD.
Recent studies in animals have demonstrated that exercise training can result in favorable changes in brain function by increasing neurogenesis, neurotrophic factors, and neuronal survival. Some human studies have indirectly supported this notion with observations of improved cognitive function and attention in people who participate in regular physical activity and who have higher cardiovascular fitness.
Based on this and other research evidence, the long-term goal of this effort is to develop and optimize an exercise intervention involving polestriding (walking with poles) that improves functional capacity in PD. This exploratory study seeks to establish that an exercise program for individuals with PD can provide improvements in functional outcomes such as balance, gait, rigidity, tremor, cardiorespiratory, and cognitive task performance. Importantly, the study will also investigate any changes in brain metabolism (glucose uptake) that may underlie such improvements. Periodic measurements of brain metabolism will be obtained using fluorodeoxyglucose positron emission tomography (18F-FDG PET) imaging that can provide a functional map of brain-wide metabolic activity. FDG-PET imaging data will be analyzed to compare any longitudinal changes in absolute and relative metabolic activities of many brain regions due to exercise. This initial investigation of exercise-induced changes in brain metabolism after PD will lay the foundation for follow-up studies to investigate and optimize exercise interventions to promote adaptation in the brain for treatment of PD.
Improving orthostatic tolerance after spinal cord injury
Key Personnel: Narayanan Krishnamurthi, PhD (PI); ANS/ASU
James J. Abbas, PhD (ANS/Bioengineering)
Sponsor: Paralyzed Veterans of America
cord injury (SCI) above the mid-thoracic region (T6) can significantly
affect autonomous nervous system function, which can lead to cardiovascular
impairments such as low resting blood pressure, orthostatic hypotension
(intolerance to change in posture from horizontal to sitting or standing
position), and blunted cardiovascular responses to exercise. Orthostatic
hypotension (OH) during either the acute or chronic stages after injury can
severely restrict participation in rehabilitation programs, thus delaying or
preventing functional recovery. Even
standard mobilization (e.g., sitting or standing) during physiotherapy is reported
to induce blood pressure reductions that produce a diagnosis of OH in 74% of
SCI patients and cause symptoms of OH in 59% of SCI individuals. Further, the
options for pharmacological treatment are limited and unpredictable due to
widely varying mechanisms of development of OH in SCI patients resulting in
greater potential for side effects. The mechanisms underlying OH are likely
to be multifactorial such as lack of tonic sympathetic control, impaired baroreceptor
regulation, lack of skeletal muscle pump activity, cardiovascular deconditioning,
and hypovolemia. Recent studies have shown that electrical stimulation
(ES)-induced contraction of leg muscles can significantly reduce or even prevent
OH by increasing the leg muscle pump activity. Most of these studies
investigated only the acute effects of the ES and it is not known if
a regular ES exercise regimen can produce persistent improvements in cardiovascular
The objective of this study is to assess the effectiveness of an ES protocol
which is safe, inexpensive and readily self-administered in a home setting. A
group of subjects with cervical or high-thoracic spinal cord injury will be
recruited to participate in a 12 week intervention that will use ES to generate
contractions of lower extremity muscles while seated. The ES parameters (stimulation
characteristics and muscles to be stimulated) will be customized for each individual
to attenuate the blood pressure drop induced by transition from a supine to
seated posture. Cardiovascular control will be characterized pre-intervention
and post-intervention through a battery of non-invasive tests using head-up
tilt (using a standard tilt table) to present a repeatable cardiovascular challenge. Assessments
will include measures to characterize changes in the acute response to electrical
stimulation as well as long-term changes in cardiovascular control.
We anticipate that successful completion of this study will demonstrate that
substantial improvements in cardiovascular control can be achieved through
participation in an exercise program that uses neuromuscular electrical stimulation. The
primary clinical impact of this program would be that it could accelerate progression
through neuromotor rehabilitation programs and that it could result in improved
cardiovascular health in the long-term. The technique holds great potential
to be widely adopted by individuals with a range of capabilities because it
involves only moderate exercise, is inexpensive, and can be readily administered
Behaviorally Relevant Neuronal Modification during Postembryonic Development
Key Personnel: S. Crook, PhD (PI) (ANS/Mathematics
and Statistics/School of Life Sciences)
C. Duch, PhD (School of Life Sciences)
Sponsor: National Science Foundation, CRCNS 10/1/2006-9/30/2009, $457,654
Understanding an adaptive and learning computing system such as the brain requires an understanding of the computations performed by its basic components--individual neurons. At the single neuron level, the dendritic tree provides the means for interpreting spatiotemporal activity patterns of synaptic input. However, no comprehensive framework exists for translating dendritic features into a functional architecture that can be related directly to behavior. Our goal is to develop rules for how dendritic structure and synapse distributions follow functional architecture principles that relate directly to a neuron's individual behavioral requirements. We are combining experimental and computational approaches to investigate the function of the structural and synaptic remodeling of insect motoneuron MN5 for its changing behavioral role. We are also testing Cajal's Neuron Doctrine, which envisions the neuron as the smallest functional entity, by examining the computational and functional independence of dendritic sub-domains. Another important aspect of our modeling studies is that we are investigating the roles of excitatory/inhibitory input synapse ratios and distributions in single neuron computation, which will contribute significantly to our understanding of single neuron computation.
Multiscale Modeling of the Neural Subcircuits in the Outer-Plexiform Layer of the Retina
Key Personnel: S. Baer, PhD (PI) (Mathematics and Statistics)
S. Crook, PhD (ANS/Mathematics and Statistics/School of Life Sciences)
C. Gardner, PhD (Mathematics and Statistics)
C. Ringhofer, PhD (Mathematics and Statistics)
Sponsor: National Science Foundation, Mathematical Biology 9/1/2007-8/31/2010, $300,000
The retina is part of the central nervous system and an ideal region for studying information processing in the brain. It is accessible, well documented, and the subject of research spanning the clinical, experimental, and theoretical sciences. Image processing begins in the outer-plexiform layer (OPL) of the retina, where bipolar, horizontal, and photoreceptor cells interact. In this first layer of the visual pathway, knowledge of synaptic feedback mechanisms allows for the formulation of computational models that encapsulate essential phenomenology. From the biologist's perspective, mathematical modeling allows one to disassociate the functional effects of various circuitry elements, similar to performing experiments with selective pharmacological agents, except that not all the agents one might like actually exist. From a mathematician's perspective, the availability of electrophysiological, anatomical, and molecular data provides a rare opportunity for the construction of complex multiscale models for the system. The primary goal of this research proposal is to mathematically model, in detail, the subcircuits of the OPL, capturing the spatio-temporal dynamics on two spatial scales-that of an individual synapse and of the receptive field. The biological merit of this proposal is to use the models to gain insight into two competing hypotheses for explaining synaptic feedback effects in the OPL. Simply put, are the feedback effects in the cone photoreceptor's synapse driven by electrical (ephaptic) or chemical (GABA) mechanisms, or both?
Adaptive Electrical Stimulation for Locomotor Retraining
Key Personnel: James J. Abbas, PhD (PI); ANS/Bioengineering
Ranu Jung, PhD (ANS/Bioengineering)
Richard Herman, MD (Banner Good Samaritan Medical Center)
Sponsor: NIH-National Center for Medical Rehabilitation Research
primary contract to ASU with subcontract to Banner Good Samaritan
Recent studies have indicated that functional recovery of locomotor function after spinal cord injury may be enhanced by performing repetitive stepping movements on a treadmill with a harness for partial body weight support with passive assistance provided by therapists. The putative mechanism that underlies this recovery is activity-dependent plasticity of neural circuits both in the spinal cord and in supraspinal centers. Although results in some subjects have been encouraging, in general, the functional gains that have been demonstrated from locomotor therapy are moderate and there is a high variability across subjects. We believe that the ‘standard’ form of this therapy (treadmill/harness with passive assistance from therapists) is soundly based on well-established principles of motor learning, but the manner in which the therapy is delivered does not enable maximization of the therapeutic effect.
We propose that locomotor therapy may be enhanced by: 1) producing sensorimotor patterns that are more ‘physiological’ - i.e. that include appropriately timed muscle contractions and are therefore more similar to sensorimotor patterns in the intact state and 2) generating movement patterns in a more repeatable manner. Our approach utilizes adaptive control of electrical stimulation to activate muscles in order to generate repeatable movements on the treadmill. We believe that the combination of appropriately-timed contractions and repeatable movement patterns will result in an improved form of locomotor therapy. Furthermore, the adaptive nature of the control system may be used to encourage gradual increases in voluntary input, therefore providing a mechanism for weaning the individual from FES-assistance during locomotion.
The long-term goal of this work is to develop a system that will provide a more effective and efficient form of locomotor retraining therapy. In this work, we will develop a technique that uses adaptive control of electrically-stimulated muscles to produce repeatable stepping movements with coordinated sensorimotor patterns of activity. The system will use transcutaneous neuromuscular stimulation to assist in movement generation while walking on the treadmill with partial body weight support provided by a harness. Adaptive control techniques will be used to automatically determine an appropriate set of stimulation parameters for a given individual and to automatically adjust the stimulation parameters to account for fatigue and/or motor retraining effects. The goals of the proposed project are to develop the adaptive system and to evaluate its ability to generate specified movement patterns. We will implement the adaptive system and experimentally demonstrate that it is capable of reliably producing stepping movements by individuals with spinal cord injury on a treadmill with partial body weight support. In future work (beyond the scope of this proposal), we will compare the efficacy of adaptive FES-assisted locomotor therapy with other forms of locomotor therapy.
CRCNS - Modeling Neuromusculoskeletal Alterations after Spinal Cord Injury
Key Personnel: R. Jung, PhD (PI); ANS/Bioengineering
J.J. Abbas, PhD (ANS/Bioengineering)
A. Iarkov, PhD (ANS)
T. Hamm, PhD (Barrows Neurological Institute)
V. Booth (University of Michigan)
Sponsor: National Institutes of Health (NIBIB; NS054282-01)
8/15/05- 7/31/09, $1,314,799
The interaction between neural and musculoskeletal systems enables us to perform a variety of motor tasks, such as locomotion, in a robust and adaptable manner. Damage to one system component, e.g. traumatic spinal cord injury, can lead to long-term secondary changes in other system components due to their close interactions and their inherent plasticity. In some instances, these secondary changes may be maladaptive, and therefore result in further reduction in functional capacity; in other instances, the changes may be favorable, and therefore result in recovery of function. In this work, a series of experimental studies in uninjured and incomplete spinal cord inured (iSCI) rodents will drive the development of a detailed mathematical model of the biomechanics and neural control of the rodent hindlimb. This model will be used to investigate the role of complex interactions amongst impaired central drive, spinal reflexes and musculoskeletal changes after iSCI in the design of appropriate therapy.
Specifically, a chronic rodent thoracic contusion spinal cord injury preparation will be used to investigate the intrinsic intracellular electrophysiology of spinal motoneurons and their afferent control and the intrinsic musculoskeletal properties present after iSCI. The experimental data will guide development of a computational model with neural and dynamic musculoskeletal components. Hodgkin-Huxley type neuron representations will be used to model the local spinal neural circuits that include motoneurons, interneurons and afferents involved in specific spinal reflexes. The musculoskeletal model will incorporate experimentally-determined geometrical musculotendon paths, inertial properties, muscle fiber properties, and 3D laser scanned bony surface geometries. The comprehensive model will consequently be used to test hypotheses regarding the roles of specific ionic currents, altered central drive, altered musculoskeletal properties and altered sensory reflex gain on control of limb movement after iSCI. Successful completion of the work will provide novel information that could help guide the development of efficient treatment techniques and appropriate rehabilitative therapies for enhancing functional locomotor recovery and quality of life for some of the 200,000 people currently living with spinal cord injury related mobility, employability and secondary health related limitations in the United States of America.
NeuroML: Standards and Tools for Multiscale Model Specification and Exchange
Key Personnel: Sharon Crook, PhD (PI): ANS, ASU
Co-investigator Suzanne Dietrich of ASU West
Sponsor: NIH R01 from NIMH
July 1, 2009 - July 1, 2012, $894,282
Force Modulation Training in Children with Cerebral Palsy
Key Personnel: A. Downing (PI, Pre-doctoral Fellowship)
J.J. Abbas, PhD (Mentor); ANS/Bioengineering
Sponsor: Ruth L. Kirschstein NRSA Pre-Doctoroal Fellowship
NIH-National Institutes for Neurological Disorders and Stroke
Cerebral palsy (CP) is the term used for a series of motor deficits resulting from an injury to an immature brain. The motor deficits of CP limit a child’s ability to explore his or her environment, thereby hindering intellectual and social stimulation. This can have significant implications for the child’s educational development, independence and quality of life. The long term goal of this work is to develop practical and effective therapeutic interventions to improve the ability of these children to explore their environment independently. One motor deficit that has been studied in children with CP is an inability to generate force (muscle weakness). Currently, weakness in children with CP has only been studied using maximum voluntary contractions; however, because most activities of daily living are performed at submaximal forces, these measurements may not adequately portray the motor deficit. In this work, we will characterize the ability of children with CP to regulate forces in the lower extremity with a particular focus on force modulation, timing, and coordination. Furthermore, we will develop and evaluate a novel force modulation training protocol intended to improve both single and multi-joint control in an attempt to produce functional gains during locomotion.
Catalyst- Center of Excellence for Adaptive Neuro-Biomechatronic Systems (CEANS)
Key Personnel: R. Jung, PhD (PI); ANS/Bioengineering
J.J. Abbas, PhD (ANS/Bioengineering)
Sharon Crook, PhD (Math/SOLS)
Carlos- Chavez-Castillo, PhD (Math)
Anthony Garcia, PhD (Bioengineering)
Lokesh Joshi, PhD( Biodesign/Bioengineering)
Yung Kuang, PhD (Math)
Anshuman Razdan, PhD (PRISM)
Stephen Phillips, PhD (Electrical Engineering)
Marco Santello, PhD (Kinesiology)
Joseph Wang, PhD (Biodesign/Chemistry)
Sponsor: National Science Foundation (Science of Learning Centers Program)
8/15/2005-7/31/2006 , $133,118
When using a tool or a device, a person must learn to interact with it appropriately in order to accomplish the task at hand. We learn to swing a hammer in a manner that strikes the nail accurately; we learn to press the brake with an appropriate level of force to decelerate the car smoothly; and we learn to move a computer mouse in order to proficiently utilize software. As technology becomes increasingly complex in its operations, its functionality, and its degree of interaction with the user, there is a growing need to embed adaptability and intelligence into the device itself. In this situation, the device and the user simultaneously learn in an attempt to optimize their interactions. The degree of success in this sensorimotor learning process depends strongly on the interaction between the person and the engineered system.
The focus of the Center of Excellence in Adaptive and Neuro-Biomechatronic Systems (CEANS) will be to understand and to optimize the learning that occurs as humans interact with adaptive technology. The Center will focus its efforts on addressing a few broad questions: What are the biological processes that occur as a person learns to interact with a device? How do the properties of the device affect the learning process? How can we design engineering devices to maximize the effectiveness of the learning process?
CEANS will use advanced prosthetic systems as a research platform to address these questions regarding the nature of the sensorimotor learning process and to develop strategies for the design of adaptive systems to achieve specific learning outcomes.
After a traumatic injury such as limb loss or spinal cord injury, technology can assist in tasks such as stepping, reaching or grasping. Several intelligent robotic devices and electrical stimulation systems with adaptive capabilities that are either on the market or on the horizon require that the person learn to use the device as the device adapts to meet the needs of the person. CEANS will use these types of co-adaptive prosthetic systems to investigate the molecular, cellular and systems-level dynamics of sensorimotor learning.
Our goal for this Catalyst project is to lay the foundation for a Science of Learning Center that addresses key issues regarding learning in the integration of adaptive biological systems with adaptive engineered systems. In the planning period, we will develop a proposal for a Center that integrates interdisciplinary research and development with educational and outreach programs.
Our research agenda will be at the intersection of molecular biology, neuroscience, mathematics, bioengineering, and rehabilitation. CEANS will draw upon a wide range of expertise to discover the principles that govern activity-dependent learning in living systems, to develop novel approaches to sense dynamic changes in adaptive living systems, and to deliver new adaptive technology for sensorimotor learning. The scope of activities will include experimental biological investigation, design and development of new technology to maximize learning outcomes, the evaluation of the effects of the technology on biological learning processes, and the transfer of these techniques to biomedical industry and clinical practice.
Hormonal Response to Exercise after Spinal Cord Injury
Key Personnel: J.J. Abbas, PhD (ANS/Bioengineering)
The long term goal of this work is to enable people with spinal cord injury to exercise in a manner that maximizes the health benefits. In this work, we seek to gain a better understanding of the physiological responses to exercise in individuals with cervical level spinal cord injury.
Humans respond to stress and exercise by activating two neuroendocrine pathways, the hypothalamus-pituitary-adrenal (HPA) axis and the sympathoadrenal (SA) system. The precisely coordinated activation of these two pathways is responsible for many of the benefits associated with exercise. A complete cervical level spinal cord injury disrupts the coordination and control of these two neuroendocrine pathways. For many individuals with spinal cord injury, the two primary options for exercise participation are arm crank ergometry, which uses voluntary contractions of upper extremity muscles, and leg cycle ergometry, which uses electrical stimulation of paralyzed muscles in the legs.
We hypothesize that the neuroendocrine response of individuals with spinal cord injury will differ for these two forms of exercise. Voluntary contractions during arm crank exercise are expected to more strongly activate the HPA axis, while electrically-induced contractions during leg cycle exercise are expected to more strongly activate the SA system.
To test this hypothesis we will recruit six subjects with complete cervical spinal cord injuries to participate in two separate exercise sessions. Session 1 will be a 30 minute bout of leg cycle exercise that uses electrically-induced contractions. Session 2 will be a 30 minute bout of voluntary arm crank exercise. We will take blood samples before and after each bout of exercise to measure changes in cortisol levels as an indicator of HPA axis activity and changes in epinephrine and norepinephrine as indicators of SA system activity. We will also measure the following metabolic markers: growth hormone, insulin, glucogon, glucose, and lactate, and the following immune markers: IL-1β, IL-2, IL-2R, IL-6, TNF-α, and CRP to further characterize the differential effects of these two forms of exercise on metabolic and immune function.
Promoting Plasticity after Spinal Cord Injury using Neuromuscular Stimulation
Key Personnel: Ranu Jung, PhD (PI); ANS/Bioengineering
James J. Abbas, PhD (ANS/Bioengineering)
Seung-Jae Kim, PhD (ANS)
Alex Iarkov, Phd (ANS)
Sponsor: Science Foundation Arizona (Competitive Advantage Award)
Approximately 250,000 individuals with spinal cord injury (SCI)
currently live in the US and approximately 11,000 people acquire new spinal injuries each year. The injuries leave people completely or partially paralyzed. There is a strong international effort for finding a cure for paralysis and many approaches trying to ameliorate the effects of spinal cord injury are being explored. In the proposed work, the neuroprosthetic technology (adaptive algorithms and neural interfaces using functional neuromuscular stimulation) will be used to provide sensorimotor rehabilitation therapy in a rodent model of spinal cord injury. Successful outcome will identify mechanisms that could be specifically targeted by the interventional therapies to promote sensorimotor recovery after incomplete spinal cord injury. The award is specifically directed towards collection of preliminary data.
Identifying novel therapeutic targets for spinal cord injury: Injury and rehabilitation mediated alterations in protein expression throughout the nervous system.
Key Personnel: J. Lynskey, PhD, PT (PI)
R. Jung, PhD (ANS)
A. Iarkov , PhD (ANS)
S-J. Kim, PhD (ANS)
Sponsor: National Institutes of Health (NIH-NICHD) via subcontract on R24 grant to Georgetown University
Development of effect therapies for spinal cord injury requires a detailed understanding of the cellular /molecular mechanisms underlying both the injury process and therapeutic agent. Rehabilitative strategies, such as locomotor training and functional electrical stimulation have been shown to promote recovery after incomplete spinal cord injury. The mechanisms and substrates mediating the effects of rehabilitation, however are poorly understood and likely involve multiple processes occurring both at the injury site and in distant regions of the neuraxis.
The proposed experiments seek to profile protein expression after incomplete spinal cord injury, with or without functional neuromuscular stimulation therapy. In this initial pilot project conducted on rodents with contusion injury , tissue will be sampled from multiple neural regions known to be involved in motor control, including the motor cortex, cerebellum, brainstem, and spinal cord.
The results will yield a detailed spatial profile of protein expression in the intact, injured and stimulated nervous system. Significant injury induced alterations in protein expression profile throughout the neuraxis, which change as a function of location and time are expected. Furthermore, FNS therapy may also produce significant alterations in protein expression profile; possibly ameliorating lesion induced aberrant changes. The data gathered from these proposed experiments should increase our understanding of spinal cord injury and neuromuscular stimulation, in addition to identifying novel therapeutic targets.
7T/30 Bruker Biospec Magnetic Resonance Imaging/Spectroscopy System
(Previously PharmaScan 70/16 In-Vivo Spectroscopy/Imaging System)
Key Personnel: Ranu Jung, PhD (P.I. Arizona Sate University)
Faculty from Arizona State University
Faculty from Good Samaritan Medical Center (Eric Reiman, MD)
Faculty from Barrows Neurological Institute (Mark Preul, MD, Adrienne Scheck, MD, Jim Pipe, PhD)
Sponsor: NIH-National Center for Research Resources
With recent genetic and molecular advances, small animal (rat/mice) models of human disease have become increasingly important resources for the investigation of the underlying mechanisms of disease. Many traditional investigational approaches require sacrificing the animals for ex vivo tissue and molecular analysis. This prevents the researchers from observing in vivo the natural or perturbed evolution of the processes under study. Additionally, small animal models are becoming increasingly important test beds to investigate the ability of novel implantable miniaturized devices and biomaterials to repair, regenerate or replace the living system. Imaging on the scale of small animals offers an opportunity to noninvasively repeat investigations of biological processes in vivo in the same animal and efficiently test treatments for disease.
One approach for bioimaging is to use nuclear magnetic resonance. The ability to perform in vivo imaging and spectroscopy in small animals or large tissue samples is absent at Arizona State University. The gap is further enhanced because of a lack of such a capability in the entire Metropolitan Phoenix Valley area that is home to several excellent clinical and research medical facilities The distance to the closest facility (120 miles) is not conducive to conducting longitudinal chronic studies on large numbers of small animals to support the needs of research in the Phoenix Valley. We were recently awarded funding for a multipurpose research scanner for high resolution, fast speed, Nuclear Magnetic Resonance 2D and 3D image reconstruction and in vivo spectroscopy. Several investigators will significantly benefit from utilizing this system in their research. The applications will range from assessing effects of chemotherapy for tumors, to developing, testing, and implementing noninvasive, brain-imaging indicators of Alzheimer’s Disease (AD) in double transgenic mice containing AD genes, to assessment of CNS neuroplasticity after spinal cord injury or stroke.