Font Size

Engineering for Neurologic Rehabilitation Program (ENRP)

ENRP at the Sensory Motor Performance Program (SMPP) of the Rehabilitation Institute of Chicago (RIC) is funded by a Medical Rehabilitation Infrastructure Award from the National Institute of Child Health and Human Development (opens new window) through the National Center for Medical Rehabilitation Research (opens new window). Our program is part of a national network of research cores that provide access to collateral expertise, and focuses on access to neuroscience and engineering resources to augment rehabilitation research in the USA. Specifically, we focus on the application of neural engineering and robotic techniques to understand, diagnose, and treat neurologic disorders.

This program is built upon core facilities established in our center under the auspices of a previously funded R24 network grant. ( Restoration of Function in Neurologic Impairment)

Researchers funded by the ENRP combine the design and problem-solving skills learned in engineering with medical and biological science to identify new avenues of investigation.  Specifically, researchers in SMPP and their collaborators are combining bioinformatics (opens new window), medical imaging (opens new window), physiological signal processing, biomechanics (opens new window) and biomaterials (opens new window) to investigate rehabilitation themes, and to train and nurture the development of investigators who plan to use such engineering applications to study neurologic disorders.

ENRP CORES

1. The Robotics/Mechanics Core:

Core Directors: James Patton, PhD, Yasin Dhaher, PhD and Sandro Mussa-Ivaldi, PhD

  • Facilities under this core are equipped with state-of-the-art lower and upper extremity robotic devices. These include the Manipulandum robot, an RIC planar machine called the reaching guide, the Phantom® 3.0 (opens new window) , the Haptic Master® (opens new window), the  WAM® (opens new window) (a novel back-drivable, three-dimensional robotics system), several WREX passive robots for gravity support of the arm, the KineAssist® and the Lokomat® (opens new window) which is a motor driven exoskeleton device that employs a body weight support suspension system combined with a split-belt instrumented treadmill (Adal®) (opens new window).

  • Researchers utilize these technologies to:  (1) address basic science and engineering questions, (2) as diagnostic tools, and (3) to test therapeutic approaches to stroke and spinal cord injury.

  • Another significant resource under this core is our Motion Analysis Laboratory.  The laboratory is equipped with an eight-camera digital, real-time, motion capture system from Motion Analysis Corporation (opens new window) (MAC, Santa Rosa, CA) that is used to measure movement kinematics.  The laboratory has three AMTI (Boston, MA) (opens new window) force platforms for measuring ground reaction forces as subjects walk across them and has an sixteen-channel EMG system (Motion Lab Systems Baton Rouge, LA) for recording muscle activity during movements, and foot switches for the identification of stance phase events during walking. All of the measurement systems in the laboratory are integrated with each other to allow for the synchronized collection of data in order to provide a comprehensive overview of the particular human movement activity.

  • The Coleman Computational Biomechanics Laboratory is the focal point of computational mechanics research at ENRP. The laboratory is equipped with six workstations with high-speed dual processors and large capacity RAM and hard drives. The six computers are also configured to operate as a cluster under the linux operating system.  Hypermesh software is available to generate geometric finite element meshes from magnetic resonance images, finite element analysis program, ABAQUS (opens new window), is used in the laboratory for modeling of joint mechanics under static and dynamic conditions. The computational laboratory also has five software licenses for Software for Interactive Musculoskeletal Modeling (SIMM) (opens new window) used to create computer models of musculoskeletal structures (single/multi-joint musculoskeletal models). SIMM is an interactive graphic tool used to enhance our understanding of the musculoskeletal biomechanics through the visualization of human movement and analysis of the functional capacity of muscles in normal and pathological states.

2. The Information Technology Core:

  • The Information Technology, Computer Simulation & Signal Processing Core helps develop measurement, analysis, and simulation tools for the study of neurological impairment.  Resources in this core include telerehabilitation facilities to quantitatively evaluate passive and active range of motion, muscle strength and stiffness of spastic joints and system identification techniques to investigate human limb dynamics under controlled conditions.  Joint kinematic, kinetic and electromyogram (EMG) signals can be measured and processed through system identification techniques to investigate various neuromechanical properties in patients with neurologic disorders and in healthy controls.

  • This core facility also has the capacity to build novel electronic and mechanical systems such as telemetered collection systems, portable data loggers, exoskeleton robots, state-of-the-art EMG electrodes, advanced signal processing systems, and extensive modeling and simulation facilities.

  • For computer modeling and simulation, the core has two licenses for Mimics (opens new window), a software package which can be used to analyze computed tomography images and to generate geometric meshes from these images for use in finite element models.  The core also has four licenses of ABAQUS (opens new window) software, available for finite element analyses in various neuromechanical applications. As noted previously, five licenses are also available for the modeling software SIMM.  The companion Dynamics Pipeline (opens new window) permits dynamic simulations of the SIMM models.

  • Software is also available for analysis of magnetic resonance images (MRIs).  MRViewer (opens new window) provides a means for segmentation of image slices, e.g., for computing lesion size.  Another workstation is available for processing of functional magnetic resonance imaging by use of software distributed through the National Institutes of Health (NIH) (opens new window) .

3. The Neuroscience Core:

  • Core Directors: C.J. Heckman, PhD (opens new window) and George Hornby PT, PhD

  • Dr. Heckman directs a research laboratory working with animal models of spinal cord injury using electrophysiological, anatomical, immunohistochemical, and pharmacological approaches.  The key goal of these studies is to develop new approaches towards therapeutic strategies in human patients.  To aid this objective, reflex pathways that can be studied in human patients are also studied in animal models, including the H-reflex, stretch reflex, tendon vibration reflex and flexion/withdrawal reflex.

  • A key resource available for these studies is a 6 degrees-of-freedom robotic arm that is used to make limb movements that are similar to those applied to human subjects.  Studies of the cellular mechanisms of reflexes and inputs generated by these limb movements in various animal models of neurotrauma and neurodegeneration suggest a diverse set of new rehabilitation strategies, ranging from specific limb manipulations to new pharmacologic agents.

  • Dr. Hornby directs the Neurolocomotion Laboratory where the focus of ongoing studies is to determine how various physical and pharmacological interventions alter volitional motor control in individuals with spinal cord injury and stroke.  He utilizes quantitative physiological, biomechanical and clinical measures to determine the mechanisms underlying alteration in motor function through recovery and rehabilitation.  Techniques employed include those utilized in the animal models, including H-, stretch, vibration and flexor withdrawal reflexes, corticomotoneuronal integrity using transcranial magnetic stimulation, as well as single and mutltijoint torques during static and dynamic conditions.  Electromyographic activity and kinematic patterns are collected during and following volitional and reflex activity, with an emphasis on locomotor function.  Cardiopulmonary measures, such as oxygen consumption and electrocardiographic activity, also help elucidate the extent of physiological improvements during our interventions.

  • In summary, our infrastructure can be considered to support a novel approach towards career development and an idea factory for rehabilitation scientists.  Thus, we have established collaborations with investigators around the country.

CURRENT COLLABORATIVE PROJECTS:

Sunil Agarwal, PhD,Steven Cramer, MD (opens new window) , Randy Powers PhD,Brian Schmit (opens new window) and Ying Wu, PhD (opens new window) have all joined with SMPP under the ENRP to form and expand our national network of researchers engineering solutions in neurologic rehabilitation for the temporary and permanently disabled.
  • Support from this award mechanism has resulted in foundational work on the design of gait rehabilitation exoskeletons and preliminary subject training. Dr. Agrawal has developed a swing-assist passive exoskeleton with the Robotics/Mechanics core of the ENRP. This work is providing new pathways and insights into future studies.  Another device, the Gravity Balancing Orthosis (GBO) has also undergone evaluation.  The results from these studies show a spectacular improvement in the range of motion of the joints, symmetry of the gait, weight bearing on the affected leg of a stroke victim, see figure below.

Related Publications and/or Presentations:

  1. S. K. Agrawal, S. Banala, A. Fattah, V. Sangwan, V. Krishnamoorthy, J. P. Scholz, W. L. Hsu, Assessment of Motion of a Swing Leg and Gait Rehabilitation with a Gravity Balancing Exoskeleton (opens new window),  IEEE Trans. on Neural Systems and Rehabilitation Engineering, Vol. 15, No. 3, 2007, 410-420).
  2. Kalyan Mankala, S. K. Agrawal, “Passive Swing Assistive Exoskeletons for Motor-Incomplete Spinal Cord Injury Patients”, In Proceedings IEEE International Conference on Robotics and Automation, Rome, April 2007.

Resulting Grant Submissions:

  1. Development and Evaluation of an Active Leg Exoskeleton for Gait Retraining after a Stroke, an NIH R01 submitted October 5, 2007.
  2. Robotic Exoskeletons, FES, and Biomechanics: Treating Movement Disorder, an NIH R01 grant proposal submitted November 5, 2007.
  3. An Upper Extremity Wearable Exoskeleton for Assistance and Training submitted to the National Science Foundation (Research to Aid Disabilities (RAPD) Program) (opens new window)  on September 15, 2007.
    • This research, supported through the ENRP, has enabled new directions of research in design and evaluation of upper and lower arm exoskeletons for training of motor impaired persons with stroke. ENRP has resulted in creation of a network of researchers to facilitate exchange of ideas and the development of next generation cutting-edge tools and methods.
  • 2.     Steven Cramer (opens new window) and his team at the University of California at Irvine (opens new window)
    • Steven Cramer and his team designed a hand-wrist assisting robotic device (HWARD) to help stroke patients improve their grasp and release abilities.  The design follows the neurobiological principles of motor learning – sensorimotor integration, environmental complexity, attention and movement repetition.  With support from the Engineering for Neurologic Rehabilitation Program and collaborative efforts with the Robotics/Mechanics core, Cramer’s group upgraded the HWARD robot including improvements to the user interface, software and clinical program.  There are now two modules, one emphasizing simple motor behaviors associated with primary motor cortex function and one emphasizing complex motor behaviors associated with premotor cortex function.

    • In this figure, (A) Study design. (B) Subject’s posture, relationship to robot and relationship to computer monitor. (C) A subject’s hand in the robot. Arrowheads indicate the splint upon which the ulnar forearm rested.White asterisks indicate the three straps connecting the hand to the robot. (D) Example of virtual reality game, the ‘jewel match’ game, which required a subject to transfer different-shaped jewels from one rotating wheel to another. As the subject’s hand moved (long arrow), the virtual hand on the computer screen (short arrow) moved, in real time (C. D. Takahashi et al. Brain (2008).

 

 

Related Publications and/or Presentations:

  1. C.D. Takahashi, L. Dcr-Yeghiaian, V.H. Le, S.C. Cramer, A Robotic Device for Hand Motor Therapy After Stroke, Proceedings of the 2005 IEEE, 9th Int’l Conference on Rehabilitation Robotics, pg 17 -20, June 28-July 1, 2005S.
  2. C.D. Takahashi, L. Der-Yeghiaian, V. Le, R. Motiwala, and S. Cramer, Robot-based hand motor therapy after stroke (opens new window) , Brain. (in press).

Resulting Grant Submissions:

  • Over the next 12 months, Dr. Cramer’s group will evaluate the improved both modules in the HWARD in up to 20 with stroke patients to develop pilot data for application of funding from the NIH to study the efficacy of robotic hand rehabilitation for stroke victims.

The current project aims to examine how content of robotic therapy might be related to the stroke lesion location.  We expect that this approach will increase the effect size of robotic therapy for motor deficits after stroke.  Such developments are expected to support collaboration and application of robotic therapy nationwide. 

3.    Randy Powers of the University of Washington School of Medicine (opens new window)

  • Randy Rowers is working on a collaborative pilot project to develop and maintain a realistic compartmental model of the motoneuron in order to test hypotheses in the biophysical properties of motoneurons that may contribute to changes in motoneuron behavior following stroke.  Dr Powers and his team have developed a compartmental model that is capable of reproducing a number of features of normal motoneuron behavior and implemented this model using NEURON simulation language. Our hypothesis is that some of the deficits in motoneuron behavior following stroke result from alterations in voltage-sensitive channels mediating persistent inward currents (PICs) carried by sodium (Na) and calcium (Ca) ions. Two drugs that inhibit Na and Ca persistent inward currents are presently in use for therapy in human subjects.  Riluzole, now a standard treatment for ALS that is classically considered to reduce glutamate release, has now been demonstrated to block Na persistent inward currents at low micromolar concentrations. Isradipine, a blood pressure medicine that crosses the blood brain barrier, blocks the Ca persistent inward current at concentrations know to have minimal side effects.  Both of these agents are presently under study by Dr. Rymer and colleagues for control of spasticity in human spinal cord injured patients. A primary advantage of our approach is that we will be able to make accurate predictions of the impact of these agents on firing rate patterns in stroke subjects, thus providing a clear guide for a potentially important and effective therapeutic strategy.
  • Alterations in other intrinsic motoneuron properties may also contribute to the deficits in motor unit behavior in stroke.  One common feature of motor unit discharge in stroke is ‘negative rate modulation’, which is a decline in firing rate of an early recruited motor unit even as the firing rate of other motor units and whole muscle force is increasing. We have recently been able to replicate this behavior in a model by slow activation of a calcium-dependent potassium conductance known to exist on rat motoneuron dendrites. The figure below shows the change in firing rate (red trace) produced in  a motoneuron model by a triangular waveform of net excitatory synaptic conductance (black trace).   The initial, rapid increase in firing rate, which reflects the activation of the Ca PIC on the dendrite, is  followed by a decline in firing rate in spite of the continued increase in net excitation, reflecting outward current flow through calcium-activated potassium channels on the dendrites.  This firing rate profile is similar to that observed in low threshold units in spastic muscles. 

  • This figure illustrates the negative rate modulation in a compartmental motoneuron model.  Instantaneous firing rate (red trace) in response to a slowly increasing then decreasing excitatory synaptic conductance input (black trace).  The initial acceleration of firing rate reflects CaPIC activation, whereas the subsequent decline reflects slow activation of a dendritic calcium-activated potassium conductance.

 

 

 

Related Publications and/or Presentations:

  • The key features and behavior of the model were briefly described in a poster presentation at the 2006 Society for Neuroscience meeting that compared to behavior in the decerebrate cat (Cope et al., 2006).

Resulting Grant Submissions :

  1. An NIH R01 submitted June, 2007
  2. An NIH R01 submitted July, 2007.
  • Brian Schmit is completing a longitudinal study of diffusion tensor imaging (DTI) in a rat model of spinal cord injury (SCI) with the Information Technology Core of the ENRP.  To date, Brian and his team have developed a fuzzy inference system for tissue classification in the human spinal cord, characterized DTI parameters in the human spinal cord and DTI imaging of the spinal cord in a rat model of injury and developed imaging sequences for magnetization transfer contrast (MTC) imaging of the spinal cord.  Future directions from these studies include expanding DTI imaging of the injured rat spinal cord.  The proposed study will examine the time-course of changes in diffusion parameters after injury and during healing.  This is part of a long-term plan to predict functional recovery from SCI using DTI measurements.

  • A longitudinal study was conducted using DTI measurements in a contusion model of the rat at the T8 segment.  Six groups of animals were tested (n= 7-12 in each group), including uninjured controls and 2, 5, 15, 20 and 25 weeks post injury.  Ex vivo images, obtained in the upper cervical spinal cord, using a 9.4T small animal MRI indicated substantial decreases in diffusivity distant from the injury, suggesting that DTI is sensitive to edema and other post-injury processes that cannot be detected using standard T2 weighted images (upper row).  Note that the longitudinal apparent diffusion coefficient (lADC) and transverse apparent diffusion coefficient (tADC) both decrease (turn blue), beginning 2 weeks post injury.  The fractional anisotropy (FA) changes only slightly, late in injury.

We are continuing to pursue a number of research directions related to DTI of the spinal cord.  First, we are looking to expand on our initial results in the rat model of SCI that suggest that DTI can be used to detect and monitor detailed characteristics of a spinal injury.  In particular, we are looking to establish the sensitivity of DTI to different severity of injury. 

Our initial work has also suggested that DTI can be used to monitor the progression of the response to injury.  As a result, we are currently preparing a study to monitor DTI of the spinal cord in two clinical conditions:

  1. in acute traumatic SCI to determine whether DTI can be used for prognosis of functional recovery and
  2. in patients with radiculopathy, to aid in prognosis and treatment planning.  A grant proposal for examining the use of DTI for SCI prognosis is currently being prepared.  We are also in the midst of preparing to translate this research to the clinic.  We are working with the Radiology Department at a local hospital in Milwaukee to routinely implement DTI scans in all acute traumatic SCI.  If successful, we anticipate that this technique will achieve widespread use.

Resulting Publications and/or Presentations:

  1. Ellingson B.M., Ulmer J.L. and Schmit B.D.  “A new technique for imaging the human spinal cord in vivo.” (opens new window) , Biomedical Sciences Instrumentation, Vol. 42, pp. 255-60, 2006.
  2. Ellingson, B.M., Ulmer, J.L. and Schmit, B.D. “Gray and white matter delineation in the human spinal cord using diffusion tensor imaging and fuzzy logic” (opens new window) , Academic Radiology, Vol 14, pp. 847-58, 2007
  3. Ellingson, B.M., Ulmer, J.L. and Schmit, B.D.  “Optimal diffusion tensor indices for imaging the human spinal cord” (opens new window), Biomedical Sciences Instrumentation. Vol. 43, pp. 128-33, 2007.
  4. Ellingson, B.M., Kurpad, S.N., Li, S.-J. and Schmit, B.D. “In vivo diffusion tensor imaging of the rat spinal cord at 9.4T” (opens new window), Journal of Magnetic Resonance Imaging, in Press.
  5. Ellingson, B.M., Ulmer, J.L., and Schmit, B.D. “Morphology of the human spinal cord in chronic injury: diffusion characteristics and tissue classification using diffusion tensor magnetic resonance imaging and fuzzy logic” (opens new window),  Annals of Biomedical Engineering, In Press
  6. Ellingson, B.M., Ulmer, J.L., Kurpad, S.N. and Schmit, B.D., “Diffusion tensor magnetic resonance imaging of the neurologically intact human spinal cord”, (Submitted)
  7. Ellingson, B.M., Schmit, B.D., Ulmer, J.L. and Kurpad, S.N., “Diffusion tensor magnetic resonance imaging in spinal cord injury”, (Submitted) 

Resulting Grant Submissions:

  1. “Diffusion Tensor Imaging of the Injured Spinal Cord” research grant submitted to the VA Merit Review (6-15-06)
  2. “Diffusion Tensor Imaging of Spinal Cord Injury”, Pre-Doctoral NRSA, (Benjamin Ellingson) submitted to the NIH (12-5-06)
  3. “Diffusion Tensor Imaging in Spinal Cord Injury”, Research Grant submitted to the Craig H. Neilsen Foundation (1-12-07)
  4. “RERC in Spinal Cord Injury: Motor Adaptations in Spinal Cord Injury”, RERC (Rehabilitation Engineering Research Center) submitted to NIDRR (3-16-07)
  5. “Diffusion Tensor Imaging of the Injured Spinal Cord” research grant submitted to the VA Merit Review (12-15-07)
  • 5.  Ying Wu (opens new window)  of Northwestern University, Department of Electrical Engineering and Computer Science is  also working with IT core of the ENRP and Dr. Rymer 
    • Ying Wu of Northwestern University, Department of Electrical Engineering and Computer Science (opens new window) is also working with IT core of the ENRP and Dr. Rymer on a collaborative pilot project titled, “Vision-based Articulated Finger Motion Analysis for Hand Rehabilitation.”  Thus far, an integrated software platform has been built, a 3-D hand model is in place for rendering, animation and analysis, essential image analysis modules are ready, a new method for estimating the articulation of the hand is being tested using synthesized and real video. We believe vision-based motion analysis can be very useful in rehabilitation in a number of application scenarios. We plan to investigate such additional applications of the developed technology, as well as, the development of new vision-based motion analysis technologies.

Related Publications and/or Presentations:

  1. Vision-based Articulated Finger Motion Analysis for Hand Rehabilitation, IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), 2008

Resulting Grant Submissions:

  1. “Video-based Non-invasive Measurement of Hand Articulation for Rehabilitation,” research grant submitted to National Science Foundation (NSF) RAPD program (2-28-2008)
  • We have been pursuing several new research directions incorporating video-based hand motion estimation and capturing. First, we are currently investigating a novel distributed and collaborative mechanism for estimating articulated hand motion. In this new approach, we have a network of correlated local motion estimators, each of which estimates a subset of the hand articulation. These local estimators are correlated. Through exchanging messages, their collaboration is expected to given computationally efficient solution to the estimation of the hand articulation. Second, we plan to develop a multiple camera system that gives multiple view image observations for motion recovery. Third, we plan to develop a method for tracking the articulated hand motion in continuous video. By taking advantage of the dynamics of the hand articulation, this will allow more accurate and more rapid motion recovery. We are submitting a grant proposal to National Science Foundation. We are currently improving the prototype video-based motion recovery system that we built, and expect more exciting results in real scenarios. We plan to submit a grant proposal to NIH in the near future.
  •  
  • For more information about the ENRP program or to collaborate with members of the ENRP, SMPP or RIC, please contact the faculty advisor for the ENRP, Yasin Dhaher, PhD
    Sensory Motor Performance Program
 

Page Updated Monday, April 28, 2008