Dr. Daniel Ferris

Invariant hip moment pattern while walking with a robotic hip exoskeleton.
Lewis CL, Ferris DP
Human Adaptation Laboratory, College of Health and Rehabilitation Sciences, Sargent College, Boston University, 635 Commonwealth Avenue, Boston, MA 02215, USA. lewisc@bu.edu
15 Mar 2011
 Abstract
Robotic lower limb exoskeletons hold significant potential for gait assistance and rehabilitation; however, we have a limited understanding of how people adapt to walking with robotic devices. The purpose of this study was to test the hypothesis that people reduce net muscle moments about their joints when robotic assistance is provided. This reduction in muscle moment results in a total joint moment (muscle plus exoskeleton) that is the same as the moment without the robotic assistance despite potential differences in joint angles. To test this hypothesis, eight healthy subjects trained with the robotic hip exoskeleton while walking on a force-measuring treadmill. The exoskeleton provided hip flexion assistance from approximately 33% to 53% of the gait cycle. We calculated the root mean squared difference (RMSD) between the average of data from the last 15 min of the powered condition and the unpowered condition. After completing three 30-min training sessions, the hip exoskeleton provided 27% of the total peak hip flexion moment during gait. Despite this substantial contribution from the exoskeleton, subjects walked with a total hip moment pattern (muscle plus exoskeleton) that was almost identical and more similar to the unpowered condition than the hip angle pattern (hip moment RMSD 0.027, angle RMSD 0.134, p<0.001). The angle and moment RMSD were not different for the knee and ankle joints. These findings support the concept that people adopt walking patterns with similar joint moment patterns despite differences in hip joint angles for a given walking speed.

Visual evoked responses during standing and walking.
Gramann K, Gwin JT, Bigdely-Shamlo N, Ferris DP, Makeig S
Swartz Center for Computational Neuroscience, Institute for Neural Computation, University of California San Diego La Jolla, CA, USA.
2010
 Abstract
Human cognition has been shaped both by our body structure and by its complex interactions with its environment. Our cognition is thus inextricably linked to our own and others' motor behavior. To model brain activity associated with natural cognition, we propose recording the concurrent brain dynamics and body movements of human subjects performing normal actions. Here we tested the feasibility of such a mobile brain/body (MoBI) imaging approach by recording high-density electroencephalographic (EEG) activity and body movements of subjects standing or walking on a treadmill while performing a visual oddball response task. Independent component analysis of the EEG data revealed visual event-related potentials that during standing, slow walking, and fast walking did not differ across movement conditions, demonstrating the viability of recording brain activity accompanying cognitive processes during whole body movement. Non-invasive and relatively low-cost MoBI studies of normal, motivated actions might improve understanding of interactions between brain and body dynamics leading to more complete biological models of cognition.

Computer simulations of neural mechanisms explaining upper and lower limb excitatory neural coupling.
Huang HJ, Ferris DP
Department of Biomedical Engineering, Human Neuromechanics Laboratory, University of Michigan, 401 Washtenaw Ave., Ann Arbor, MI 48109-2214, USA. helen.huang@colorado.edu
2010
 Abstract
When humans perform rhythmic upper and lower limb locomotor-like movements, there is an excitatory effect of upper limb exertion on lower limb muscle recruitment. To investigate potential neural mechanisms for this behavioral observation, we developed computer simulations modeling interlimb neural pathways among central pattern generators. We hypothesized that enhancement of muscle recruitment from interlimb spinal mechanisms was not sufficient to explain muscle enhancement levels observed in experimental data.We used Matsuoka oscillators for the central pattern generators (CPG) and determined parameters that enhanced amplitudes of rhythmic steady state bursts. Potential mechanisms for output enhancement were excitatory and inhibitory sensory feedback gains, excitatory and inhibitory interlimb coupling gains, and coupling geometry. We first simulated the simplest case, a single CPG, and then expanded the model to have two CPGs and lastly four CPGs. In the two and four CPG models, the lower limb CPGs did not receive supraspinal input such that the only mechanisms available for enhancing output were interlimb coupling gains and sensory feedback gains.In a two-CPG model with inhibitory sensory feedback gains, only excitatory gains of ipsilateral flexor-extensor/extensor-flexor coupling produced reciprocal upper-lower limb bursts and enhanced output up to 26%. In a two-CPG model with excitatory sensory feedback gains, excitatory gains of contralateral flexor-flexor/extensor-extensor coupling produced reciprocal upper-lower limb bursts and enhanced output up to 100%. However, within a given excitatory sensory feedback gain, enhancement due to excitatory interlimb gains could only reach levels up to 20%. Interconnecting four CPGs to have ipsilateral flexor-extensor/extensor-flexor coupling, contralateral flexor-flexor/extensor-extensor coupling, and bilateral flexor-extensor/extensor-flexor coupling could enhance motor output up to 32%. Enhancement observed in experimental data exceeded 32%. Enhancement within this symmetrical four-CPG neural architecture was more sensitive to relatively small interlimb coupling gains. Excitatory sensory feedback gains could produce greater output amplitudes, but larger gains were required for entrainment compared to inhibitory sensory feedback gains.Based on these simulations, symmetrical interlimb coupling can account for much, but not all of the excitatory neural coupling between upper and lower limbs during rhythmic locomotor-like movements.

The effects of error augmentation on learning to walk on a narrow balance beam.
Domingo A, Ferris DP
School of Kinesiology, University of Michigan, 3158 Observatory Lodge, 1402 Washtenaw Avenue, Ann Arbor, MI 48109-2214, USA. antoinette.domingo@ubc.ca
Oct 2010
 Abstract
Error augmentation during training has been proposed as a means to facilitate motor learning due to the human nervous system's reliance on performance errors to shape motor commands. We studied the effects of error augmentation on short-term learning of walking on a balance beam to determine whether it had beneficial effects on motor performance. Four groups of able-bodied subjects walked on a treadmill-mounted balance beam (2.5-cm wide) before and after 30 min of training. During training, two groups walked on the beam with a destabilization device that augmented error (Medium and High Destabilization groups). A third group walked on a narrower beam (1.27-cm) to augment error (Narrow). The fourth group practiced walking on the 2.5-cm balance beam (Wide). Subjects in the Wide group had significantly greater improvements after training than the error augmentation groups. The High Destabilization group had significantly less performance gains than the Narrow group in spite of similar failures per minute during training. In a follow-up experiment, a fifth group of subjects (Assisted) practiced with a device that greatly reduced catastrophic errors (i.e., stepping off the beam) but maintained similar pelvic movement variability. Performance gains were significantly greater in the Wide group than the Assisted group, indicating that catastrophic errors were important for short-term learning. We conclude that increasing errors during practice via destabilization and a narrower balance beam did not improve short-term learning of beam walking. In addition, the presence of qualitatively catastrophic errors seems to improve short-term learning of walking balance.

Electrocortical activity is coupled to gait cycle phase during treadmill walking.
Gwin JT, Gramann K, Makeig S, Ferris DP
Human Neuromechanics Laboratory, School of Kinesiology, University of Michigan, Ann Arbor, MI 48109-2214, USA. jgwin@umich.edu
15 Jan 2011
 Abstract
Recent findings suggest that human cortex is more active during steady-speed unperturbed locomotion than previously thought. However, techniques that have been used to image the brain during locomotion lack the temporal resolution necessary to assess intra-stride cortical dynamics. Our aim was to determine if electrocortical activity is coupled to gait cycle phase during steady-speed human walking. We used electroencephalography (EEG), motion capture, and a force-measuring treadmill to record brain and body dynamics while eight healthy young adult subjects walked on a treadmill. Infomax independent component analysis (ICA) parsed EEG signals into maximally independent component (IC) processes representing electrocortical sources, muscle sources, and artifacts. We calculated a spatially fixed equivalent current dipole for each IC using an inverse modeling approach, and clustered electrocortical sources across subjects by similarities in dipole locations and power spectra. We then computed spectrograms for each electrocortical source that were time-locked to the gait cycle. Electrocortical sources in the anterior cingulate, posterior parietal, and sensorimotor cortex exhibited significant (p<0.05) intra-stride changes in spectral power. During the end of stance, as the leading foot was contacting the ground and the trailing foot was pushing off, alpha- and beta-band spectral power increased in or near the left/right sensorimotor and dorsal anterior cingulate cortex. Power increases in the left/right sensorimotor cortex were more pronounced for contralateral limb push-off (ipsilateral heel-strike) than for ipsilateral limb push-off (contralateral heel-strike). Intra-stride high-gamma spectral power changes were evident in anterior cingulate, posterior parietal, and sensorimotor cortex. These data confirm cortical involvement in steady-speed human locomotion. Future applications of these techniques could provide critical insight into the neural mechanisms of movement disorders and gait rehabilitation.

Short-term locomotor adaptation to a robotic ankle exoskeleton does not alter soleus Hoffmann reflex amplitude.
Kao PC, Lewis CL, Ferris DP
School of Kinesiology, University of Michigan, Ann Arbor, Michigan 48109-2214, USA. kaop@umich.edu
2010
 Abstract
To improve design of robotic lower limb exoskeletons for gait rehabilitation, it is critical to identify neural mechanisms that govern locomotor adaptation to robotic assistance. Previously, we demonstrated soleus muscle recruitment decreased by approximately 35% when walking with a pneumatically-powered ankle exoskeleton providing plantar flexor torque under soleus proportional myoelectric control. Since a substantial portion of soleus activation during walking results from the stretch reflex, increased reflex inhibition is one potential mechanism for reducing soleus recruitment when walking with exoskeleton assistance. This is clinically relevant because many neurologically impaired populations have hyperactive stretch reflexes and training to reduce the reflexes could lead to substantial improvements in their motor ability. The purpose of this study was to quantify soleus Hoffmann (H-) reflex responses during powered versus unpowered walking.We tested soleus H-reflex responses in neurologically intact subjects (n=8) that had trained walking with the soleus controlled robotic ankle exoskeleton. Soleus H-reflex was tested at the mid and late stance while subjects walked with the exoskeleton on the treadmill at 1.25 m/s, first without power (first unpowered), then with power (powered), and finally without power again (second unpowered). We also collected joint kinematics and electromyography.When the robotic plantar flexor torque was provided, subjects walked with lower soleus electromyographic (EMG) activation (27-48%) and had concomitant reductions in H-reflex amplitude (12-24%) compared to the first unpowered condition. The H-reflex amplitude in proportion to the background soleus EMG during powered walking was not significantly different from the two unpowered conditions.These findings suggest that the nervous system does not inhibit the soleus H-reflex in response to short-term adaption to exoskeleton assistance. Future studies should determine if the findings also apply to long-term adaption to the exoskeleton.

Removal of movement artifact from high-density EEG recorded during walking and running.
Gwin JT, Gramann K, Makeig S, Ferris DP
Human Neuromechanics Laboratory, School of Kinesiology; University of Michigan, Ann Arbor, Michigan, USA. jgwin@umich.edu
Jun 2010
 Abstract
Although human cognition often occurs during dynamic motor actions, most studies of human brain dynamics examine subjects in static seated or prone conditions. EEG signals have historically been considered to be too noise prone to allow recording of brain dynamics during human locomotion. Here we applied a channel-based artifact template regression procedure and a subsequent spatial filtering approach to remove gait-related movement artifact from EEG signals recorded during walking and running. We first used stride time warping to remove gait artifact from high-density EEG recorded during a visual oddball discrimination task performed while walking and running. Next, we applied infomax independent component analysis (ICA) to parse the channel-based noise reduced EEG signals into maximally independent components (ICs) and then performed component-based template regression. Applying channel-based or channel-based plus component-based artifact rejection significantly reduced EEG spectral power in the 1.5- to 8.5-Hz frequency range during walking and running. In walking conditions, gait-related artifact was insubstantial: event-related potentials (ERPs), which were nearly identical to visual oddball discrimination events while standing, were visible before and after applying noise reduction. In the running condition, gait-related artifact severely compromised the EEG signals: stable average ERP time-courses of IC processes were only detectable after artifact removal. These findings show that high-density EEG can be used to study brain dynamics during whole body movements and that mechanical artifact from rhythmic gait events may be minimized using a template regression procedure.

Joint kinetic response during unexpectedly reduced plantar flexor torque provided by a robotic ankle exoskeleton during walking.
Kao PC, Lewis CL, Ferris DP
School of Kinesiology, University of Michigan, Ann Arbor, MI 48109-2214, USA. kaop@udel.edu
7 May 2010
 Abstract
During human walking, plantar flexor activation in late stance helps to generate a stable and economical gait pattern. Because plantar flexor activation is highly mediated by proprioceptive feedback, the nervous system must modulate reflex pathways to meet the mechanical requirements of gait. The purpose of this study was to quantify ankle joint mechanical output of the plantar flexor stretch reflex response during a novel unexpected gait perturbation. We used a robotic ankle exoskeleton to mechanically amplify the ankle torque output resulting from soleus muscle activation. We recorded lower-body kinematics, ground reaction forces, and electromyography during steady-state walking and during randomly perturbed steps when the exoskeleton assistance was unexpectedly turned off. We also measured soleus Hoffmann- (H-) reflexes at late stance during the two conditions. Subjects reacted to the unexpectedly decreased exoskeleton assistance by greatly increasing soleus muscle activity about 60ms after ankle angle deviated from the control condition (p<0.001). There were large differences in ankle kinematic and electromyography patterns for the perturbed and control steps, but the total ankle moment was almost identical for the two conditions (p=0.13). The ratio of soleus H-reflex amplitude to background electromyography was not significantly different between the two conditions (p=0.4). This is the first study to show that the nervous system chooses reflex responses during human walking such that invariant ankle joint moment patterns are maintained during perturbations. Our findings are particularly useful for the development of neuromusculoskeletal computer simulations of human walking that need to adjust reflex gains appropriately for biomechanical analyses.

Slacking by the human motor system: computational models and implications for robotic orthoses.
Reinkensmeyer DJ, Akoner O, Ferris DP, Gordon KE
Department of Mechanical and Aerospace Engineering and Department of Biomedical Engineering, University of California at Irvine, CA 92617, USA. dreinken@uci.edu
2009
 Abstract
Recent experimental evidence suggests that a fundamental property of the human motor system is that it "slacks"; that is, that it continuously attempts to decrease levels of muscle activation when movement error is small during repetitive motions. This paper reviews several computational models of slacking, and discusses implications of slacking for the design of robotic orthoses. For therapeutic applications of robotic orthoses, slacking may reduce human effort during rehabilitation training, with negative consequences for use-dependent motor recovery. For assistive applications of robotic orthoses, slacking may allow the motor system to learn to take advantage of force amplification provided by an orthosis, with positive consequences for human energy efficiency.

Robotic lower limb exoskeletons using proportional myoelectric control.
Ferris DP, Lewis CL
School of Kinesiology, University of Michigan, Ann Arbor, MI 48109-2013 USA. ferrisdp@umich.edu
2009
 Abstract
Robotic lower limb exoskeletons have been built for augmenting human performance, assisting with disabilities, studying human physiology, and re-training motor deficiencies. At the University of Michigan Human Neuromechanics Laboratory, we have built pneumatically-powered lower limb exoskeletons for the last two purposes. Most of our prior research has focused on ankle joint exoskeletons because of the large contribution from plantar flexors to the mechanical work performed during gait. One way we control the exoskeletons is with proportional myoelectric control, effectively increasing the strength of the wearer with a physiological mode of control. Healthy human subjects quickly adapt to walking with the robotic ankle exoskeletons, reducing their overall energy expenditure. Individuals with incomplete spinal cord injury have demonstrated rapid modification of muscle recruitment patterns with practice walking with the ankle exoskeletons. Evidence suggests that proportional myoelectric control may have distinct advantages over other types of control for robotic exoskeletons in basic science and rehabilitation.

Preliminary trial of symmetry-based resistance in individuals with post-stroke hemiparesis.
Simon AM, Kelly BM, Ferris DP
Department of Biomedical Engineering, University of Michigan, Ann Arbor, MI 48109, USA. annie-simon@northwestern.edu
2009
 Abstract
We tested a novel control strategy for robotic rehabilitation devices used by individuals with post-stroke hemiparesis. Symmetry-based resistance increases resistance when limb forces become more asymmetric during bilateral exercise. The underlying rationales for the control mode are that it will guide patients to increase paretic limb activation while teaching them to accurately gauge paretic limb force production relative to the non-paretic limb. During a one day training session, seven subjects post-stroke performed lower limb extensions in symmetry-based resistance mode on a robotic exercise machine. Subjects improved lower limb symmetry from 28.6%+/-3.9% to 36.2%+/-4.3% while under symmetry-based resistance training (ANOVA, P = 0.03), but did not maintain the improved lower limb symmetry during a constant resistance post-test. Two subjects that showed the large improvements in symmetry during the one day session performed additional days of training. Those results suggest that some patients demonstrate long lasting benefits with symmetry-based resistance training.

Invariant ankle moment patterns when walking with and without a robotic ankle exoskeleton.
Kao PC, Lewis CL, Ferris DP
School of Kinesiology, University of Michigan, Ann Arbor, MI 48109-2214, USA. kaop@udel.edu
19 Jan 2010
 Abstract
To guide development of robotic lower limb exoskeletons, it is necessary to understand how humans adapt to powered assistance. The purposes of this study were to quantify joint moments while healthy subjects adapted to a robotic ankle exoskeleton and to determine if the period of motor adaptation is dependent on the magnitude of robotic assistance. The pneumatically powered ankle exoskeleton provided plantar flexor torque controlled by the wearer's soleus electromyography (EMG). Eleven naïve individuals completed two 30-min sessions walking on a split-belt instrumented treadmill at 1.25m/s while wearing the ankle exoskeleton. After two sessions of practice, subjects reduced their soleus EMG activation by approximately 36% and walked with total ankle moment patterns similar to their unassisted gait (r(2)=0.98+/-0.02, THSD, p>0.05). They had substantially different ankle kinematic patterns compared to their unassisted gait (r(2)=0.79+/-0.12, THSD, p<0.05). Not all of the subjects reached a steady-state gait pattern within the two sessions, in contrast to a previous study using a weaker robotic ankle exoskeleton (Gordon and Ferris, 2007). Our results strongly suggest that humans aim for similar joint moment patterns when walking with robotic assistance rather than similar kinematic patterns. In addition, greater robotic assistance provided during initial use results in a longer adaptation process than lesser robotic assistance.

Upper limb effort does not increase maximal voluntary muscle activation in individuals with incomplete spinal cord injury.
Huang HJ, Ferris DP
Department of Biomedical Engineering, University of Michigan, Ann Arbor, MI, USA. helen.huang@colorado.edu
Sep 2009
 Abstract
To determine the effect of upper limb effort on maximal lower limb muscle activation in individuals with incomplete spinal cord injury.Fifteen individuals with incomplete spinal cord injury performed recumbent stepping using different combinations of upper and lower limb efforts.There was no significant difference in active lower limb electromyography amplitudes regardless of whether the upper limbs were resting or exerting maximal effort. Upper limb effort increased passive lower limb muscle activation and likewise, lower limb effort increased passive upper limb muscle activation.Upper limb effort did not increase lower limb muscle activation during active lower limb effort in individuals with incomplete spinal cord injury during recumbent stepping. This suggests that individuals with incomplete spinal cord injury cannot recruit additional lower limb motor units using maximal volitional effort of their upper limbs.Understanding how upper limb effort and movement influences lower limb muscle activation patterns in incomplete spinal cord injury patients has implications for prescribing therapies for lower limb rehabilitation.

Effects of physical guidance on short-term learning of walking on a narrow beam.
Domingo A, Ferris DP
School of Kinesiology, University of Michigan, Ann Arbor, MI 48109-2214, USA. adomingo@umich.edu
Nov 2009
 Abstract
Physical guidance is often used in rehabilitation when teaching patients to re-learn movements. However, the effects of guidance on motor learning of complex skills, such as walking balance, are not clear. We tested four groups of healthy subjects that practiced walking on a narrow (1.27 cm) or wide (2.5 cm) treadmill-mounted balance beam, with or without physical guidance. Assistance was given by springs attached to a hip belt that applied restoring forces towards beam center. Subjects were evaluated while walking unassisted before and after training by calculating the number of times subjects stepped off of the beam per minute of successful walking on the beam (Failures per Minute). Subjects in Unassisted groups had greater performance improvements in walking balance from pre to post compared to subjects in Assisted groups. During training, Unassisted groups had more Failures per Minute than Assisted groups. Performance improvements were smaller in Narrow Beam groups than in Wide Beam groups. The Unassisted-Wide and Assisted-Narrow groups had similar Failures per Minute during training, but the Unassisted-Wide group had much greater performance gains after training. These results suggest that physical assistance can hinder motor learning of walking balance, assistance appears less detrimental for more difficult tasks, and task-specific dynamics are important to learning independent of error experience.

Upper and lower limb muscle activation is bidirectionally and ipsilaterally coupled.
Huang HJ, Ferris DP
Department of Biomedical Engineering, University of Michigan, Ann Arbor, MI, USA. helen.huang@colorado.edu
Sep 2009
 Abstract
There are neural connections between the upper and lower limbs of humans that enable muscle activation in one limb pair (upper or lower) to modulate muscle activation in the other limb pair (lower or upper, respectively). The aims of this study were to extend previous findings regarding submaximal exercise to maximal effort exercise and determine whether there is an ipsilateral or contralateral bias to the neural coupling during a rhythmic locomotor-like task.We measured upper and lower limb muscle activity, joint kinematics, and limb forces in neurologically intact subjects (n = 16) as they performed recumbent stepping using different combinations of upper and lower limb efforts.We found increased muscle activation in passive lower limbs during active upper limb effort compared with passive upper limb effort. Likewise, increased muscle activation in passive upper limbs occurred during active lower limb effort compared with passive lower limb effort, suggesting a bidirectional effect. Maximal muscle activation in the active lower limbs was not different between conditions with active upper limb effort and conditions with passive upper limb movement. Similarly, maximal muscle activation in the active upper limbs was not different between conditions with active lower limb effort and conditions with passive lower limb movement. Further comparisons revealed that neural coupling was primarily from active upper limb muscles to passive ipsilateral lower limb muscles.These findings indicate that interlimb neural coupling affects muscle recruitment during maximal effort upper and lower limb rhythmic exercise and provides insight into the architecture of the neural coupling.

Neuromechanical considerations for incorporating rhythmic arm movement in the rehabilitation of walking.
Klimstra MD, Thomas E, Stoloff RH, Ferris DP, Zehr EP
Rehabilitation Neuroscience Laboratory, University of Victoria, Victoria, British Columbia V8W 3P1, Canada.
Jun 2009
 Abstract
We have extensively used arm cycling to study the neural control of rhythmic movements such as arm swing during walking. Recently rhythmic movement of the arms has also been shown to enhance and shape muscle activity in the legs. However, restricted information is available concerning the conditions necessary to maximally alter lumbar spinal cord excitability. Knowledge on the neuromechanics of a task can assist in the determination of the type, level, and timing of neural signals, yet arm swing during walking and arm cycling have not received a detailed neuromechanical comparison. The purpose of this research was to provide a combined neural and mechanical measurement approach that could be used to assist in the determination of the necessary and sufficient conditions for arm movement to assist in lower limb rehabilitation after stroke and spinal cord injury. Subjects performed three rhythmic arm movement tasks: (1) cycling (cycle); (2) swinging while standing (swing); and (3) swinging while treadmill walking (walk). We hypothesized that any difference in neural control between tasks (i.e., pattern of muscle activity) would reflect changes in the mechanical constraints unique to each task. Three-dimensional kinematics were collected simultaneously with force measurement at the hand and electromyography from the arms and trunk. All data were appropriately segmented to allow a comparison between and across conditions and were normalized and averaged to 100% movement cycle based on shoulder excursion. Separate mathematical principal components analysis of kinematic and neural variables was performed to determine common task features and muscle synergies. The results highlight important neural and mechanical features that distinguish differences between tasks. For example, there are considerable differences in the anatomical positions of the arms during each task, which relate to the moments experienced about the elbow and shoulder. Also, there are differences between tasks in elbow flexion/extension kinematics alongside differential muscle activation profiles. As well, mechanical assistance and constraints during all tasks could affect muscle recruitment and the functional role of muscles. Overall, despite neural and mechanical differences, the results are consistent with conserved common central motor control mechanisms operational for cycle, walk, and swing but appropriately sculpted to demands unique to each task. However, changing the mechanical parameters could affect the role of afferent feedback altering neural control and the coupling to the lower limbs.

It pays to have a spring in your step.
Sawicki GS, Lewis CL, Ferris DP
Department of Ecology and Evolutionary Biology, Brown University, Providence, RI, USA.
Jul 2009
 Abstract
In humans, a large portion of the mechanical work required for walking comes from muscle-tendons crossing the ankle joint. Elastic energy storage and return in the Achilles tendon during each step enhance the efficiency of ankle muscle-tendon mechanical work far beyond what is possible for work performed by knee and hip joint muscle-tendons.

A pneumatically powered knee-ankle-foot orthosis (KAFO) with myoelectric activation and inhibition.
Sawicki GS, Ferris DP
Human Neuromechanics Laboratory, School of Kinesiology, University of Michigan, 401 Washtenaw Avenue, Ann Arbor, Michigan 48109-2214, USA. gsawicki@umich.edu
2009
 Abstract
The goal of this study was to test the mechanical performance of a prototype knee-ankle-foot orthosis (KAFO) powered by artificial pneumatic muscles during human walking. We had previously built a powered ankle-foot orthosis (AFO) and used it effectively in studies on human motor adaptation, locomotion energetics, and gait rehabilitation. Extending the previous AFO to a KAFO presented additional challenges related to the force-length properties of the artificial pneumatic muscles and the presence of multiple antagonistic artificial pneumatic muscle pairs.Three healthy males were fitted with custom KAFOs equipped with artificial pneumatic muscles to power ankle plantar flexion/dorsiflexion and knee extension/flexion. Subjects walked over ground at 1.25 m/s under four conditions without extensive practice: 1) without wearing the orthosis, 2) wearing the orthosis with artificial muscles turned off, 3) wearing the orthosis activated under direct proportional myoelectric control, and 4) wearing the orthosis activated under proportional myoelectric control with flexor inhibition produced by leg extensor muscle activation. We collected joint kinematics, ground reaction forces, electromyography, and orthosis kinetics.The KAFO produced approximately 22%-33% of the peak knee flexor moment, approximately 15%-33% of the peak extensor moment, approximately 42%-46% of the peak plantar flexor moment, and approximately 83%-129% of the peak dorsiflexor moment during normal walking. With flexor inhibition produced by leg extensor muscle activation, ankle (Pearson r-value = 0.74 +/- 0.04) and knee ( r = 0.95 +/- 0.04) joint kinematic profiles were more similar to the without orthosis condition compared to when there was no flexor inhibition (r = 0.49 +/- 0.13 for ankle, p = 0.05, and r = 0.90 +/- 0.03 for knee, p = 0.17).The proportional myoelectric control with flexor inhibition allowed for a more normal gait than direct proportional myoelectric control. The current orthosis design provided knee torques smaller than the ankle torques due to the trade-off in torque and range of motion that occurs with artificial pneumatic muscles. Future KAFO designs could incorporate cams, gears, or different actuators to transmit greater torque to the knee.

The exoskeletons are here.
Ferris DP
School of Kinesiology, University of Michigan, Ann Arbor, MI, USA. ferrisdp@umich.edu
2009
 Abstract
It is a fantastic time for the field of robotic exoskeletons. Recent advances in actuators, sensors, materials, batteries, and computer processors have given new hope to creating the exoskeletons of yesteryear's science fiction. While the most common goal of an exoskeleton is to provide superhuman strength or endurance, scientists and engineers around the world are building exoskeletons with a wide range of diverse purposes. Exoskeletons can help patients with neurological disabilities improve their motor performance by providing task specific practice. Exoskeletons can help physiologists better understand how the human body works by providing a novel experimental perturbation. Exoskeletons can even help power mobile phones, music players, and other portable electronic devices by siphoning mechanical work performed during human locomotion. This special thematic series on robotic lower limb exoskeletons and orthoses includes eight papers presenting novel contributions to the field. The collective message of the papers is that robotic exoskeletons will contribute in many ways to the future benefit of humankind, and that future is not that distant.

Sense of effort determines lower limb force production during dynamic movement in individuals with poststroke hemiparesis.
Simon AM, Kelly BM, Ferris DP
Department of Biomedical Engineering, University of Michigan, Ann Arbor, Michigan. asimon@umich.edu
Oct 2009
 Abstract
This study's purpose was to determine if individuals who have had a stroke primarily use sense of effort to gauge force production during static and dynamic lower limb contractions. If relying on sense of effort while attempting to generate equal limb forces, participants should produce equal percentages of their maximum voluntary strength rather than equal absolute forces in their limbs.Ten stroke participants performed isometric and isotonic lower limb extensions on an exercise machine.When participants attempted to produce equal bilateral isometric forces, there was a significant difference in absolute force between limbs (ANOVA, P < .0001) but no significant difference when force was normalized to each limb's maximum voluntary contraction (MVC) force (P = .5129). During bilateral isotonic contractions, participants produced less absolute force in their paretic limb (P = .0005) and less relative force in their paretic limb (normalized to MVC force) when participants were given no instructions on how to perform the extension (P = .0002). When participants were instructed to produce equal forces, there was no significant difference between relative forces in the 2 limbs (P = .2111).For both isometric and isotonic conditions hemiparetic participants relied primarily on sense of effort, rather than proprioceptive feedback, for gauging lower limb force production. This outcome indicates that sense of effort is the major factor determining force production during movements. Lower limb rehabilitation therapies should not only train strength in the paretic limb but should also train patients to recalibrate force-scaling abilities to improve function.

Medial gastrocnemius myoelectric control of a robotic ankle exoskeleton.
Kinnaird CR, Ferris DP
Division of Kinesiology, Ann Arbor, MI 48109, USA
Feb 2009
 Abstract
A previous study from our laboratory showed that when soleus electromyography was used to control the amount of plantar flexion assistance from a robotic ankle exoskeleton, subjects significantly reduced their soleus activity to quickly return to normal gait kinematics. We speculated that subjects were primarily responding to the local mechanical assistance of the exoskeleton rather than directly attempting to reduce exoskeleton mechanical power via decreases in soleus activity. To test this observation we studied ten healthy subjects walking on a treadmill at 1.25 m/s while wearing a robotic exoskeleton proportionally controlled by medial gastrocnemius activation. We hypothesized that subjects would primarily decrease soleus activity due to its synergistic mechanics with the exoskeleton. Subjects decreased medial gastrocnemius recruitment by 12% ( p < 0.05 ) but decreased soleus recruitment by 27% ( p < 0.05). In agreement with our hypothesis, the primary reduction in muscle activity was not for the control muscle (medial gastrocnemius) but for the anatomical synergist to the exoskeleton (soleus). These findings indicate that anatomical morphology needs to be considered carefully when designing software and hardware for robotic exoskeletons.

Metabolic and mechanical energy costs of reducing vertical center of mass movement during gait.
Gordon KE, Ferris DP, Kuo AD
Department of Movement Science, University of Michigan, Ann Arbor, MI, USA. keith-gordon@northwestern.edu
Jan 2009
 Abstract
To test the hypothesis that reducing vertical center of mass (COM) displacement will lower the metabolic cost of human walking. To examine changes in joint work associated with increasing and decreasing vertical COM movement during gait.Randomized repeated measures.Human Neuromechanics Laboratory, University of Michigan.Able-bodied subjects (N=10).Subjects walked at 1.2m/s on a treadmill and overground. Subjects manipulated vertical COM displacement either by adjusting stride length or by using visual feedback to reduce COM movement.We measured kinematic and kinetic data to calculate vertical and lateral COM displacements, joint torques, and work. In addition, we collected oxygen consumption to calculated metabolic power.Increasing and decreasing vertical COM displacement beyond subjects' preferred range resulted in increases in the metabolic cost of walking. When vertical COM displacement was reduced, corresponding increases in positive ankle and hip work and negative knee work were observed.Humans are capable of walking in a manner that will reduce COM displacement from normal. Decreasing vertical COM movement results in increases in metabolic energy costs because of greater mechanical work performed at the hip, knee, and ankle joints. Thus, reducing vertical COM movement is not a successful strategy for improving either metabolic or mechanical energy economy during normal walking by able-bodied subjects.

Mechanics and energetics of incline walking with robotic ankle exoskeletons.
Sawicki GS, Ferris DP
Human Neuromechanics Laboratory, University of Michigan-Ann Arbor, Ann Arbor, MI 48109, USA. gsawicki@brown.edu
Jan 2009
 Abstract
We examined healthy human subjects wearing robotic ankle exoskeletons to study the metabolic cost of ankle muscle-tendon work during uphill walking. The exoskeletons were powered by artificial pneumatic muscles and controlled by the user's soleus electromyography. We hypothesized that as the demand for net positive external mechanical work increased with surface gradient, the positive work delivered by ankle exoskeletons would produce greater reductions in users' metabolic cost. Nine human subjects walked at 1.25 m s(-1) on gradients of 0%, 5%, 10% and 15%. We compared rates of O(2) consumption and CO(2) production, exoskeleton mechanics, joint kinematics, and surface electromyography between unpowered and powered exoskeleton conditions. On steeper inclines, ankle exoskeletons delivered more average positive mechanical power (P<0.0001; +0.37+/-0.03 W kg(-1) at 15% grade and +0.23+/-0.02 W kg(-1) at 0% grade) and reduced subjects' net metabolic power by more (P<0.0001; -0.98+/-0.12 W kg(-1) at 15% grade and -0.45+/-0.07 W kg(-1) at 0% grade). Soleus muscle activity was reduced by 16-25% when wearing powered exoskeletons on all surface gradients (P<0.0008). The ;apparent efficiency' of ankle muscle-tendon mechanical work decreased from 0.53 on level ground to 0.38 on 15% grade. This suggests a decreased contribution from previously stored Achilles' tendon elastic energy and an increased contribution from actively shortening ankle plantar flexor muscle fibers to ankle muscle-tendon positive work during walking on steep uphill inclines. Although exoskeletons delivered 61% more mechanical work at the ankle up a 15% grade compared with level walking, relative reductions in net metabolic power were similar across surface gradients (10-13%). These results suggest a shift in the relative distribution of mechanical power output to more proximal (knee and hip) joints during inclined walking.

Powered ankle exoskeletons reveal the metabolic cost of plantar flexor mechanical work during walking with longer steps at constant step frequency.
Sawicki GS, Ferris DP
Human Neuromechanics Laboratory, University of Michigan at Ann Arbor, Ann Arbor, MI 48109, USA. gsawicki@brown.edu
Jan 2009
 Abstract
We examined the metabolic cost of plantar flexor muscle-tendon mechanical work during human walking. Nine healthy subjects walked at constant step frequency on a motorized treadmill at speeds corresponding to 80% (1.00 m s(-1)), 100% (1.25 m s(-1)), 120% (1.50 m s(-1)) and 140% (1.75 m s(-1)) of their preferred step length (L(*)) at 1.25 m s(-1). In each condition subjects donned robotic ankle exoskeletons on both legs. The exoskeletons were powered by artificial pneumatic muscles and controlled using soleus electromyography (i.e. proportional myoelectric control). We measured subjects' metabolic energy expenditure and exoskeleton mechanics during both unpowered and powered walking to test the hypothesis that ankle plantarflexion requires more net metabolic power (W kg(-1)) at longer step lengths for a constant step frequency (i.e. preferred at 1.25 m s(-1)). As step length increased from 0.8 L(*) to 1.4 L(*), exoskeletons delivered approximately 25% more average positive mechanical power (P=0.01; +0.20+/-0.02 W kg(-1) to +0.25+/-0.02 W kg(-1), respectively). The exoskeletons reduced net metabolic power by more at longer step lengths (P=0.002; -0.21+/-0.06 W kg(-1) at 0.8 L(*) and -0.70+/-0.12 W kg(-1) at 1.4 L(*)). For every 1 J of exoskeleton positive mechanical work subjects saved 0.72 J of metabolic energy ('apparent efficiency'=1.39) at 0.8 L(*) and 2.6 J of metabolic energy ('apparent efficiency'=0.38) at 1.4 L(*). Declining ankle muscle-tendon ;apparent efficiency' suggests an increase in ankle plantar flexor muscle work relative to Achilles' tendon elastic energy recoil during walking with longer steps. However, previously stored elastic energy in Achilles' tendon still probably contributes up to 34% of ankle muscle-tendon positive work even at the longest step lengths we tested. Across the range of step lengths we studied, the human ankle muscle-tendon system performed 34-40% of the total lower-limb positive mechanical work but accounted for only 7-26% of the net metabolic cost of walking.

Motor adaptation during dorsiflexion-assisted walking with a powered orthosis.
Kao PC, Ferris DP
Division of Kinesiology, 401 Washtenaw Avenue, University of Michigan, Ann Arbor, MI 48109-2214, USA. kaop@umich.edu
Feb 2009
 Abstract
A robotic ankle-foot orthosis (AFO) that provides powered assistance could adjust to varying gait dynamics much better than a rigid AFO. To provide insight into how humans would adapt to a powered AFO, we studied the response of neurologically intact subjects walking with an active dorsiflexion assist orthosis proportionally controlled by tibialis anterior electromyography (EMG). We examined the two mechanical functions of ankle dorsiflexors in gait (power absorption at heel strike and power generation at toe-off) by recruiting two groups of healthy subjects: Group One, called Continuous Control (n=5), had dorsiflexion assistance both at the initial heel contact and during swing; Group Two, called Swing Control (n=5), had the assistance only during swing. We hypothesized both groups of subjects would reduce tibialis anterior EMG amplitude with practice walking with the powered dorsiflexion assist. Ten healthy subjects were fitted with custom-made orthoses that included an artificial pneumatic muscle providing dorsiflexor torque. We collected lower body kinematics, EMG, and artificial muscle force while subjects walked on a treadmill for two 30-min training sessions. We found that subjects walked with increased ankle dorsiflexion by 9 degrees but showed different adaptation responses of the two tibialis anterior EMG bursts. The first EMG burst around heel strike had approximately 28% lower amplitudes (p<0.05) but the second EMG burst during swing had similar amplitudes. These results provide baseline data of EMG controlled dorsiflexion assist in neurologically intact humans that can be used to guide future studies on neurologically impaired individuals.

A simple method for calibrating force plates and force treadmills using an instrumented pole.
Collins SH, Adamczyk PG, Ferris DP, Kuo AD
Department of Mechanical Engineering, University of Michigan, Ann Arbor, MI, USA. shc@umich.edu
Jan 2009
 Abstract
We propose a new method for calibrating force plates to reduce errors in center of pressure locations, forces, and moments. These errors may be caused by imperfect mounting of force plates to the ground or by installation of a treadmill atop a force plate, which may introduce distorting loads. The method, termed the Post-Installation Least-Squares (PILS) calibration, combines features of several previous methods into a simple procedure. It requires a motion capture system and an instrumented pole for applying reference loads. Reference loads may be manually applied to the force plate in arbitrary locations and directions. The instrumented pole measures applied load magnitudes through a single-axis load cell, and load directions through motion capture markers. Reference data and imperfect force plate signals are then combined to form a linear calibration matrix that simultaneously minimizes mean square errors in all forces and moments. We applied the procedure to standard laboratory force plates, as well as a custom-built, split-belt force treadmill. We also collected an independent set of verification data for testing. The proposed calibration procedure was found to reduce force errors by over 20%, and moment errors by over 60%. Center of pressure errors were also reduced by 63% for standard force plates and 91% for the force treadmill. The instrumented pole is advantageous because it allows for fast and arbitrary load application without needing a precise fixture for aligning loads. The linear calibration matrix is simpler than nonlinear correction equations and more compatible with standard data acquisition software, yet it yields error reductions comparable to more complex methods.

Walking with increased ankle pushoff decreases hip muscle moments.
Lewis CL, Ferris DP
Division of Kinesiology, University of Michigan, 401 Washtenaw Avenue, Ann Arbor, MI 48109-2214, USA. caralew@umich.edu
19 Jul 2008
 Abstract
In a simple bipedal walking model, an impulsive push along the trailing limb (similar to ankle plantar flexion) or a torque at the hip can power level walking. This suggests a tradeoff between ankle and hip muscle requirements during human gait. People with anterior hip pain may benefit from walking with increased ankle pushoff if it reduces hip muscle forces. The purpose of our study was to determine if simple instructions to alter ankle pushoff can modify gait dynamics and if resulting changes in ankle pushoff have an effect on hip muscle requirements during gait. We hypothesized that changes in ankle kinetics would be inversely related to hip muscle kinetics. Ten healthy subjects walked on a custom split-belt force-measuring treadmill at 1.25m/s. We recorded ground reaction forces and lower extremity kinematic data to calculate joint angles and internal muscle moments, powers and angular impulses. Subjects walked under three conditions: natural pushoff, decreased pushoff and increased pushoff. For the decreased pushoff condition, subjects were instructed to push less with their feet as they walked. Conversely, for the increased pushoff condition, subjects were instructed to push more with their feet. As predicted, walking with increased ankle pushoff resulted in lower peak hip flexion moment, power and angular impulse as well as lower peak hip extension moment and angular impulse (p<0.05). Our results emphasize the interchange between hip and ankle kinetics in human walking and suggest that increased ankle pushoff during gait may help to compensate for hip muscle weakness or injury and reduce hip joint forces.

Mechanics and energetics of level walking with powered ankle exoskeletons.
Sawicki GS, Ferris DP
Human Neuromechanics Laboratory, University of Michigan-Ann Arbor, Ann Arbor, MI 48109, USA.
May 2008
 Abstract
Robotic lower limb exoskeletons that can alter joint mechanical power output are novel tools for studying the relationship between the mechanics and energetics of human locomotion. We built pneumatically powered ankle exoskeletons controlled by the user's own soleus electromyography (i.e. proportional myoelectric control) to determine whether mechanical assistance at the ankle joint could reduce the metabolic cost of level, steady-speed human walking. We hypothesized that subjects would reduce their net metabolic power in proportion to the average positive mechanical power delivered by the bilateral ankle exoskeletons. Nine healthy individuals completed three 30 min sessions walking at 1.25 m s(-1) while wearing the exoskeletons. Over the three sessions, subjects' net metabolic energy expenditure during powered walking progressed from +7% to -10% of that during unpowered walking. With practice, subjects significantly reduced soleus muscle activity (by approximately 28% root mean square EMG, P<0.0001) and negative exoskeleton mechanical power (-0.09 W kg(-1) at the beginning of session 1 and -0.03 W kg(-1) at the end of session 3; P=0.005). Ankle joint kinematics returned to similar patterns to those observed during unpowered walking. At the end of the third session, the powered exoskeletons delivered approximately 63% of the average ankle joint positive mechanical power and approximately 22% of the total positive mechanical power generated by all of the joints summed (ankle, knee and hip) during unpowered walking. Decreases in total joint positive mechanical power due to powered ankle assistance ( approximately 22%) were not proportional to reductions in net metabolic power ( approximately 10%). The ;apparent efficiency' of the ankle joint muscle-tendon system during human walking ( approximately 0.61) was much greater than reported values of the ;muscular efficiency' of positive mechanical work for human muscle ( approximately 0.10-0.34). High ankle joint ;apparent efficiency' suggests that recoiling Achilles' tendon contributes a significant amount of ankle joint positive power during the push-off phase of walking in humans.

Resonant hopping of a robot controlled by an artificial neural oscillator.
Pelc EH, Daley MA, Ferris DP
Department of Biomedical Engineering, University of Michigan, Ann Arbor, MI 48109, USA.
Jun 2008
 Abstract
The bouncing gaits of terrestrial animals (hopping, running, trotting) can be modeled as a hybrid dynamic system, with spring-mass dynamics during stance and ballistic motion during the aerial phase. We used a simple hopping robot controlled by an artificial neural oscillator to test the ability of the neural oscillator to adaptively drive this hybrid dynamic system. The robot had a single joint, actuated by an artificial pneumatic muscle in series with a tendon spring. We examined how the oscillator-robot system responded to variation in two neural control parameters: descending neural drive and neuromuscular gain. We also tested the ability of the oscillator-robot system to adapt to variations in mechanical properties by changing the series and parallel spring stiffnesses. Across a 100-fold variation in both supraspinal gain and muscle gain, hopping frequency changed by less than 10%. The neural oscillator consistently drove the system at the resonant half-period for the stance phase, and adapted to a new resonant half-period when the muscle series and parallel stiffnesses were altered. Passive cycling of elastic energy in the tendon accounted for 70-79% of the mechanical work done during each hop cycle. Our results demonstrate that hopping dynamics were largely determined by the intrinsic properties of the mechanical system, not the specific choice of neural oscillator parameters. The findings provide the first evidence that an artificial neural oscillator will drive a hybrid dynamic system at partial resonance.

Lower limb force production and bilateral force asymmetries are based on sense of effort.
Simon AM, Ferris DP
Department of Biomedical Engineering, University of Michigan, 401 Washtenaw Ave., Ann Arbor, MI 48109-2214, USA. asimon@umich.edu
May 2008
 Abstract
Previous research suggests that individuals use a sense of effort, more than proprioceptive feedback, to gauge force production in their upper limbs. We have adopted an isometric force matching task to determine if force asymmetry between lower limbs during bilateral force production results from a neural mechanism related to sense of effort. We hypothesized that subjects attempting to produce equal lower limb forces would generate equal percentages of their bilateral maximum voluntary strength rather than equal absolute limb forces. Ten subjects performed isometric lower limb extensions on an exercise machine. Subjects attempted to match forces in their lower limbs at three different submaximal levels (20, 40, and 60% of their weaker limb peak force during bilateral maximum voluntary contraction). Subjects received visual feedback of only the target and stronger limb force. Results showed that subjects consistently produced less force in their weaker limb during all force matching levels when normalized to their unilateral maximum voluntary contraction force (ANOVAs 20% P = 0.0473, 40% P = 0.0012, 60% P = 0.0007). As predicted by our hypothesis, normalizing force magnitudes by bilateral maximum voluntary contraction forces revealed no significant differences between limbs at all force levels (ANOVA P = 0.8490). Regardless of whether humans produce maximal or submaximal forces, limb force asymmetry appears to be related to neural factors rather than differences in mechanical capabilities between the limbs. Our findings have implications for bilateral asymmetries during movement in healthy and neurologically impaired populations.

A PHYSIOLOGIST'S PERSPECTIVE ON ROBOTIC EXOSKELETONS FOR HUMAN LOCOMOTION.
Ferris DP, Sawicki GS, Daley MA
Human Neuromechanics Laboratory, University of Michigan, 401 Washtenaw Ave., Ann Arbor, MI 48109-2214, USA.
Sep 2007
 Abstract
Technological advances in robotic hardware and software have enabled powered exoskeletons to move from science fiction to the real world. The objective of this article is to emphasize two main points for future research. First, the design of future devices could be improved by exploiting biomechanical principles of animal locomotion. Two goals in exoskeleton research could particularly benefit from additional physiological perspective: 1) reduction in the metabolic energy expenditure of the user while wearing the device, and 2) minimization of the power requirements for actuating the exoskeleton. Second, a reciprocal potential exists for robotic exoskeletons to advance our understanding of human locomotor physiology. Experimental data from humans walking and running with robotic exoskeletons could provide important insight into the metabolic cost of locomotion that is impossible to gain with other methods. Given the mutual benefits of collaboration, it is imperative that engineers and physiologists work together in future studies on robotic exoskeletons for human locomotion.

Locomotor adaptation to a powered ankle-foot orthosis depends on control method.
Cain SM, Gordon KE, Ferris DP
Department of Biomedical Engineering, University of Michigan, 1107 Carl A, Gerstacker, 2200 Bonisteel Blvd,, Ann Arbor, MI 48109-2099, USA. smcain@umich.edu
2007
 Abstract
We studied human locomotor adaptation to powered ankle-foot orthoses with the intent of identifying differences between two different orthosis control methods. The first orthosis control method used a footswitch to provide bang-bang control (a kinematic control) and the second orthosis control method used a proportional myoelectric signal from the soleus (a physiological control). Both controllers activated an artificial pneumatic muscle providing plantar flexion torque.Subjects walked on a treadmill for two thirty-minute sessions spaced three days apart under either footswitch control (n = 6) or myoelectric control (n = 6). We recorded lower limb electromyography (EMG), joint kinematics, and orthosis kinetics. We compared stance phase EMG amplitudes, correlation of joint angle patterns, and mechanical work performed by the powered orthosis between the two controllers over time.During steady state at the end of the second session, subjects using proportional myoelectric control had much lower soleus and gastrocnemius activation than the subjects using footswitch control. The substantial decrease in triceps surae recruitment allowed the proportional myoelectric control subjects to walk with ankle kinematics close to normal and reduce negative work performed by the orthosis. The footswitch control subjects walked with substantially perturbed ankle kinematics and performed more negative work with the orthosis.These results provide evidence that the choice of orthosis control method can greatly alter how humans adapt to powered orthosis assistance during walking. Specifically, proportional myoelectric control results in larger reductions in muscle activation and gait kinematics more similar to normal compared to footswitch control.

Kinematics and muscle activity of individuals with incomplete spinal cord injury during treadmill stepping with and without manual assistance.
Domingo A, Sawicki GS, Ferris DP
Division of Kinesiology, University of Michigan, Ann Arbor, MI, USA. adomingo@umich.edu
2007
 Abstract
Treadmill training with bodyweight support and manual assistance improves walking ability of patients with neurological injury. The purpose of this study was to determine how manual assistance changes muscle activation and kinematic patterns during treadmill training in individuals with incomplete spinal cord injury.We tested six volunteers with incomplete spinal cord injury and six volunteers with intact nervous systems. Subjects with spinal cord injury walked on a treadmill at six speeds (0.18-1.07 m/s) with body weight support with and without manual assistance. Healthy subjects walked at the same speeds only with body weight support. We measured electromyographic (EMG) and kinematics in the lower extremities and calculated EMG root mean square (RMS) amplitudes and joint excursions. We performed cross-correlation analyses to compare EMG and kinematic profiles.Normalized muscle activation amplitudes and profiles in subjects with spinal cord injury were similar for stepping with and without manual assistance (ANOVA, p > 0.05). Muscle activation amplitudes increased with increasing speed (ANOVA, p < 0.05). When comparing spinal cord injury subject EMG data to control subject EMG data, neither the condition with manual assistance nor the condition without manual assistance showed a greater similarity to the control subject data, except for vastus lateralis. The shape and timing of EMG patterns in subjects with spinal cord injury became less similar to controls at faster speeds, especially when walking without manual assistance (ANOVA, p < 0.05). There were no consistent changes in kinematic profiles across spinal cord injury subjects when they were given manual assistance. Knee joint excursion was approximately 5 degrees greater with manual assistance during swing (ANOVA, p < 0.05). Hip and ankle joint excursions were both approximately 3 degrees lower with manual assistance during stance (ANOVA, p < 0.05).Providing manual assistance does not lower EMG amplitudes or alter muscle activation profiles in relatively higher functioning spinal cord injury subjects. One advantage of manual assistance is that it allows spinal cord injury subjects to walk at faster speeds than they could without assistance. Concerns that manual assistance will promote passivity in subjects are unsupported by our findings.

Neural regulation of rhythmic arm and leg movement is conserved across human locomotor tasks.
Zehr EP, Balter JE, Ferris DP, Hundza SR, Loadman PM, Stoloff RH
Rehabilitation Neuroscience Laboratory, PO Box 3010 STN CSC, University of Victoria, Victoria, BC, Canada V8W 3P1. pzehr@uvic.ca
1 Jul 2007
 Abstract
It has been proposed that different forms of rhythmic human limb movement have a common central neural control ('common core hypothesis'), just as in other animals. We compared the modulation patterns of background EMG and cutaneous reflexes during walking, arm and leg cycling, and arm-assisted recumbent stepping. We hypothesized that patterns of EMG and reflex modulation during cycling and stepping (deduced from mathematical principal components analysis) would be comparable to those during walking because they rely on similar neural substrates. Differences between the tasks were assessed by evoking cutaneous reflexes via stimulation of nerves in the foot and hand in separate trials. The EMG was recorded from flexor and extensor muscles of the arms and legs. Angular positions of the hip, knee and elbow joints were also recorded. Factor analysis revealed that across the three tasks, four principal components explained more than 93% of the variance in the background EMG and middle-latency reflex amplitude. Phase modulation of reflex amplitude was observed in most muscles across all tasks, suggesting activity in similar control networks. Significant correlations between EMG level and reflex amplitude were frequently observed only during static voluntary muscle activation and not during rhythmic movement. Results from a control experiment showed that strong correlation between EMG and reflex amplitudes was observed during discrete, voluntary leg extension but not during walking. There were task-dependent differences in reflex modulation between the three tasks which probably arise owing to specific constraints during each task. Overall, the results show strong correlation across tasks and support common neural patterning as the regulator of arm and leg movement during various rhythmic human movements.

Learning to walk with a robotic ankle exoskeleton.
Gordon KE, Ferris DP
Division of Kinesiology, University of Michigan, 401 Washtenaw Avenue, Ann Arbor, MI 48109-2214, USA. keith-gordon@northwestern.edu
2007
 Abstract
We used a lower limb robotic exoskeleton controlled by the wearer's muscle activity to study human locomotor adaptation to disrupted muscular coordination. Ten healthy subjects walked while wearing a pneumatically powered ankle exoskeleton on one limb that effectively increased plantar flexor strength of the soleus muscle. Soleus electromyography amplitude controlled plantar flexion assistance from the exoskeleton in real time. We hypothesized that subjects' gait kinematics would be initially distorted by the added exoskeleton power, but that subjects would reduce soleus muscle recruitment with practice to return to gait kinematics more similar to normal. We also examined the ability of subjects to recall their adapted motor pattern for exoskeleton walking by testing subjects on two separate sessions, 3 days apart. The mechanical power added by the exoskeleton greatly perturbed ankle joint movements at first, causing subjects to walk with significantly increased plantar flexion during stance. With practice, subjects reduced soleus recruitment by approximately 35% and learned to use the exoskeleton to perform almost exclusively positive work about the ankle. Subjects demonstrated the ability to retain the adapted locomotor pattern between testing sessions as evidenced by similar muscle activity, kinematic and kinetic patterns between the end of the first test day and the beginning of the second. These results demonstrate that robotic exoskeletons controlled by muscle activity could be useful tools for testing neural mechanisms of human locomotor adaptation.

Recumbent stepping has similar but simpler neural control compared to walking.
Stoloff RH, Zehr EP, Ferris DP
Department of Movement Science, University of Michigan, Ann Arbor, MI 48109-2214, USA.
Apr 2007
 Abstract
The purpose of this study was to compare muscle activation patterns and kinematics during recumbent stepping and walking to determine if recumbent stepping has a similar motor pattern as walking. We measured joint kinematics and electromyography in ten neurologically intact humans walking on a treadmill at 0 and 50% body weight support (BWS), and recumbent stepping using a commercially available exercise machine. Cross correlation of upper and lower limb electromyography patterns between conditions revealed high correlations for most muscles. A principal component analysis revealed that the first factor accounted for more muscle activation signal content during recumbent stepping (81%) than during walking (70%). This indicates that the motor pattern during walking is more complex than during stepping. Cross correlation analysis found a high correlation between factors for recumbent stepping and walking (R = 0.54), though not as high as the correlation between factors for walking at 0% BWS and walking at 50% BWS (R = 0.68). There were substantial differences in joint kinematics between walking and recumbent stepping, most notably in hip, elbow, and shoulder motions. These results suggest that although the two tasks have different kinematic patterns, recumbent stepping relies on similar neural networks as walking. Individuals with neurological impairments may be able to improve walking ability from recumbent stepping practice given similarities in neural control between the two tasks.

Symmetry-based resistance as a novel means of lower limb rehabilitation.
Simon AM, Brent Gillespie R, Ferris DP
Department of Biomedical Engineering, The University of Michigan, Ann Arbor, MI 48109-2214, USA. asimon@umich.edu
2007
 Abstract
Robotic devices hold much promise for use as rehabilitation aids but their success depends on identifying effective strategies for controlling human-robot interaction forces. We developed a robotic device to test a novel method of controlling interaction forces with the intent of improving force symmetry in the limbs. Users perform lower limb extensions against a computer-controlled resistive load. The control software increases resistance above baseline in proportion to lower limb force asymmetry (balance between left and right limb forces). As a preliminary trial to test the device and controller, we conducted two experiments on neurologically intact subjects. In experiment 1, one group of subjects received symmetry-based resistance while performing lower limb extensions (n=10). A control group performed the same movements with constant resistance (n=10). The symmetry-based resistance group improved lower limb symmetry during training (ANOVA, p<0.05), whereas the control subjects did not. In experiment 2, subjects (n=10) successfully used symmetry-based resistance to alter their lower limb force production towards a target asymmetry (ANOVA, p<0.05). These studies suggest that symmetry-based resistance may hold rehabilitation benefits after orthopedic or neurological injury. Specifically, performing strength training therapy with this controller may allow hemiparetic individuals to focus better on increasing strength and neuromuscular recruitment in their paretic limb while experiencing symmetric limb forces.

Moving the arms to activate the legs.
Ferris DP, Huang HJ, Kao PC
Department of Movement Science, University of Michigan, Ann Arbor, MI 48109-2214, USA. ferrisdp@umich.edu
Jul 2006
 Abstract
Recent studies on neurologically intact individuals and individuals with spinal cord injury indicate that rhythmic upper limb muscle activation has an excitatory effect on lower limb muscle activation during locomotor-like tasks. This finding suggests that gait rehabilitation therapy after neurological injury should incorporate simultaneous upper limb and lower limb rhythmic exercise to take advantage of neural coupling.

Powered lower limb orthoses for gait rehabilitation.
Ferris DP, Sawicki GS, Domingo A
2005
 Abstract
Bodyweight supported treadmill training has become a prominent gait rehabilitation method in leading rehabilitation centers. This type of locomotor training has many functional benefits but the labor costs are considerable. To reduce therapist effort, several groups have developed large robotic devices for assisting treadmill stepping. A complementary approach that has not been adequately explored is to use powered lower limb orthoses for locomotor training. Recent advances in robotic technology have made lightweight powered orthoses feasible and practical. An advantage to using powered orthoses as rehabilitation aids is they allow practice starting, turning, stopping, and avoiding obstacles during overground walking.

The effects of powered ankle-foot orthoses on joint kinematics and muscle activation during walking in individuals with incomplete spinal cord injury.
Sawicki GS, Domingo A, Ferris DP
Division of Kinesiology, University of Michigan, Ann Arbor, MI, USA. gsawicki@umich.edu
2006
 Abstract
Powered lower limb orthoses could reduce therapist labor during gait rehabilitation after neurological injury. However, it is not clear how patients respond to powered assistance during stepping. Patients might allow the orthoses to drive the movement pattern and reduce their muscle activation. The goal of this study was to test the effects of robotic assistance in subjects with incomplete spinal cord injury using pneumatically powered ankle-foot orthoses.Five individuals with chronic incomplete spinal cord injury (ASIA C-D) participated in the study. Each subject was fitted with bilateral ankle-foot orthoses equipped with artificial pneumatic muscles to power ankle plantar flexion. Subjects walked on a treadmill with partial bodyweight support at four speeds (0.36, 0.54, 0.72 and 0.89 m/s) under three conditions: without wearing orthoses, wearing orthoses unpowered (passively), and wearing orthoses activated under pushbutton control by a physical therapist. Subjects also attempted a fourth condition wearing orthoses activated under pushbutton control by them. We measured joint angles, electromyography, and orthoses torque assistance.A therapist quickly learned to activate the artificial pneumatic muscles using the pushbuttons with the appropriate amplitude and timing. The powered orthoses provided approximately 50% of peak ankle torque. Ankle angle at stance push-off increased when subjects walked with powered orthoses versus when they walked with passive-orthoses (ANOVA, p < 0.05). Ankle muscle activation amplitudes were similar for powered and passive-orthoses conditions except for the soleus (approximately 13% lower for powered condition; p < 0.05). Two of the five subjects were able to control the orthoses themselves using the pushbuttons. The other three subjects found it too difficult to coordinate pushbutton timing. Orthoses assistance and maximum ankle angle at push-off were smaller when the subject controlled the orthoses compared to when the therapist-controlled the orthoses (p < 0.05). Muscle activation amplitudes were similar between the two powered conditions except for tibialis anterior (approximately 31% lower for therapist-controlled; p < 0.05).Mechanical assistance from powered ankle-foot orthoses improved ankle push-off kinematics without substantially reducing muscle activation during walking in subjects with incomplete spinal cord injury. These results suggest that robotic plantar flexion assistance could be used during gait rehabilitation without promoting patient passivity.

Neuromechanical adaptation to hopping with an elastic ankle-foot orthosis.
Ferris DP, Bohra ZA, Lukos JR, Kinnaird CR
Division of Kinesiology, 401 Washtenaw Ave., University of Michigan, Ann Arbor, Michigan 48109-2214, USA. ferrisdp@umich.edu
Jan 2006
 Abstract
When humans hop or run on different surfaces, they adjust their effective leg stiffness to offset changes in surface stiffness. As a result, the overall stiffness of the leg-surface series combination remains independent of surface stiffness. The purpose of this study was to determine whether humans make a similar adjustment when springs are placed in parallel with the leg via a lower limb orthosis. We studied seven human subjects hopping in place on one leg while wearing an ankle-foot orthosis. We used an ankle-foot orthosis because the ankle joint is primarily responsible for leg stiffness during hopping. A spring was added to the ankle-foot orthosis so that it increased orthosis stiffness by providing plantar flexor torque during ankle dorsiflexion. We hypothesized that subjects would decrease their biological ankle stiffness when the spring was added to the orthosis, keeping total ankle stiffness constant. We collected kinematic, kinetic, and electromyographic data during hopping with and without the spring on the orthosis. We found that total ankle stiffness and leg stiffness did not change across the two orthosis conditions (ANOVA, P > 0.05). This was possible because subjects decreased their biological ankle stiffness to offset the orthosis spring stiffness (P < 0.0001). The reduction in biological ankle stiffness was accompanied by decreases in soleus, medial gastrocnemius, and lateral gastrocnemius muscle activation (P < 0.0002). These results suggest that an elastic exoskeleton might improve human running performance by reducing muscle recruitment.

An improved powered ankle-foot orthosis using proportional myoelectric control.
Ferris DP, Gordon KE, Sawicki GS, Peethambaran A
Department of Movement Science, University of Michigan, Ann Arbor, MI 48109-2214, USA. ferrisdp@umich.edu
Jun 2006
 Abstract
We constructed a powered ankle-foot orthosis for human walking with a novel myoelectric controller. The orthosis included a carbon fiber and polypropylene shell, a metal hinge joint, and two artificial pneumatic muscles. Soleus electromyography (EMG) activated the artificial plantar flexor and inhibited the artificial dorsiflexor. Tibialis anterior EMG activated the artificial dorsiflexor. We collected kinematic, kinetic, and electromyographic data for a naive healthy subject walking with the orthosis. The current design improves upon a previous prototype by being easier to don and doff and simpler to use. The novel controller allows naive wearers to quickly adapt to the orthosis without artificial muscle co-contraction. The orthosis may be helpful in studying human walking biomechanics and assisting patients during gait rehabilitation after neurological injury.

An ankle-foot orthosis powered by artificial pneumatic muscles.
Ferris DP, Czerniecki JM, Hannaford B
Department of Movement Science, University of Michigan, Ann Arbor, MI 48109-2214, USA.
May 2005
 Abstract
We developed a pneumatically powered orthosis for the human ankle joint. The orthosis consisted of a carbon fiber shell, hinge joint, and two artificial pneumatic muscles. One artificial pneumatic muscle provided plantar flexion torque and the second one provided dorsiflexion torque. Computer software adjusted air pressure in each artificial muscle independently so that artificial muscle force was proportional to rectified low-pass-filtered electromyography (EMG) amplitude (i.e., proportional myoelectric control). Tibialis anterior EMG activated the artificial dorsiflexor and soleus EMG activated the artificial plantar flexor. We collected joint kinematic and artificial muscle force data as one healthy participant walked on a treadmill with the orthosis. Peak plantar flexor torque provided by the orthosis was 70 Nm, and peak dorsiflexor torque provided by the orthosis was 38 Nm. The orthosis could be useful for basic science studies on human locomotion or possibly for gait rehabilitation after neurological injury.

Mechanical performance of artificial pneumatic muscles to power an ankle-foot orthosis.
Gordon KE, Sawicki GS, Ferris DP
Department of Movement Science, University of Michigan, 401 Washtenaw Avenue, Ann Arbor, 48109-2214, USA. kegordon@umich.edu
2006
 Abstract
We developed a powered ankle-foot orthosis that uses artificial pneumatic muscles to produce active plantar flexor torque. The purpose of this study was to quantify the mechanical performance of the orthosis during human walking. Three subjects walked at a range of speeds wearing ankle-foot orthoses with either one or two artificial muscles working in parallel. The orthosis produced similar total peak plantar flexor torque and network across speeds independent of the number of muscles used. The orthosis generated approximately 57% of the peak ankle plantar flexor torque during stance and performed approximately 70% of the positive plantar flexor work done during normal walking. Artificial muscle bandwidth and force-length properties were the two primary factors limiting torque production. The lack of peak force and work differences between single and double muscle conditions can be explained by force-length properties. Subjects altered their ankle kinematics between conditions resulting in changes in artificial muscle length. In the double muscle condition greater plantar flexion yielded shorter artificial muscles lengths and decreased muscle forces. This finding emphasizes the importance of human testing in the design and development of robotic exoskeleton devices for assisting human movement. The results of this study outline the mechanical performance limitations of an ankle-foot orthosis powered by artificial pneumatic muscles. This orthosis could be valuable for gait rehabilitation and for studies investigating neuromechanical control of human walking.

The effect of movement frequency on interlimb coupling during recumbent stepping.
Kao PC, Ferris DP
Dept of Movement Science, University of Michigan, Ann Arbor, MI, USA.
Apr 2005
 Abstract
During passive lower limb movement, active use of the upper limbs increases unintentional lower limb muscle activation. We hypothesized that faster movement frequencies would amplify lower limb muscle activation during upper limb exertion but would not affect lower limb muscle activation when the upper limbs were relaxed. We studied 10 healthy participants exercising on a recumbent stepping machine that mechanically coupled the four limbs via handles and pedals. Participants exercised at four frequencies (30, 60, 90, 120 steps/min) under four conditions of active and passive movement. Self-driven lower limb motion resulted in greater muscle activation compared to externally driven lower limb motion. Muscle activation amplitude increased with frequency for all conditions except for externally driven stepping. These results indicate that fast upper limb movement facilitates neuromuscular recruitment of lower limb muscles during stepping tasks. If a similar effect occurs in neurologically impaired individuals during active stepping, self-assisted exercise might enhance neuromuscular recruitment during rehabilitation.

Proportional myoelectric control of a virtual object to investigate human efferent control.
Gordon KE, Ferris DP
Department of Movement Science, University of Michigan, 401 Washtenaw Avenue, Ann Arbor, MI 48103-2214, USA. kegordon@umich.edu
Dec 2004
 Abstract
We used proportional myoelectric control of a one-dimensional virtual object to investigate differences in efferent control between the proximal and distal muscles of the upper limbs. Eleven subjects placed one of their upper limbs in a brace that restricted movement while we recorded electromyography (EMG) signals from elbow flexors/extensors or wrist flexors/extensors during isometric contractions. By activating their muscles, subjects applied virtual forces to a virtual object using a real-time computer interface. The magnitudes of these forces were proportional to EMG amplitudes. Subjects used this proportional EMG control to move the virtual object through two tracking tasks, one with a static target and one with a moving target (i.show $132#e., a sine wave). We hypothesized that subjects would have better control over the virtual object using their distal muscles rather than using their proximal muscles because humans typically use more distal joints to perform fine motor tasks. The results indicated that there was no difference in subjects' ability to control virtual object movements when using either upper arm muscles or forearm muscles. These results suggest that differences in control accuracy between elbow joint movements and wrist joint movements are more likely to be a result of motor practice, proprioceptive feedback or joint mechanics rather than inherent differences in efferent control.

Neural coupling between upper and lower limbs during recumbent stepping.
Huang HJ, Ferris DP
Department of Biomedical Engineering, Human Neuromechanics Laboratory, 1206A CCRB, 401 Washtenaw Ave., Univ. of Michigan, Ann Arbor, MI 48109-2214, USA. hjhuang@umich.edu
Oct 2004
 Abstract
During gait rehabilitation, therapists or robotic devices often supply physical assistance to a patient's lower limbs to aid stepping. The expensive equipment and intensive manual labor required for these therapies limit their availability to patients. One alternative solution is to design devices where patients could use their upper limbs to provide physical assistance to their lower limbs (i.e., self-assistance). To explore potential neural effects of coupling upper and lower limbs, we investigated neuromuscular recruitment during self-driven and externally driven lower limb motion. Healthy subjects exercised on a recumbent stepper using different combinations of upper and lower limb exertions. The recumbent stepper mechanically coupled the upper and lower limbs, allowing users to drive the stepping motion with upper and/or lower limbs. We instructed subjects to step with 1) active upper and lower limbs at an easy resistance level (active arms and legs); 2) active upper limbs and relaxed lower limbs at easy, medium, and hard resistance levels (self-driven); and 3) relaxed upper and lower limbs while another person drove the stepping motion (externally driven). We recorded surface electromyography (EMG) from six lower limb muscles. Self-driven EMG amplitudes were always higher than externally driven EMG amplitudes (P < 0.05). As resistance and upper limb exertion increased, self-driven EMG amplitudes also increased. EMG bursts during self-driven and active arms and legs stepping occurred at similar times. These results indicate that active upper limb movement increases neuromuscular activation of the lower limbs during cyclic stepping motions. Neurologically impaired humans that actively engage their upper limbs during gait rehabilitation may increase neuromuscular activation and enhance activity-dependent plasticity.