Benjamin D. Robertson

Elastic ankle exoskeletons reduce soleus muscle force but not work in human hopping

Inspired by elastic energy storage and return in tendons of human leg muscle-tendon units (MTU), exoskeletons often place a spring in parallel with an MTU to assist the MTU. However, this might perturb the normally efficient MTU mechanics and actually increase active muscle mechanical work. This study tested the effects of elastic parallel assistance on MTU mechanics. Participants hopped with and without spring-loaded ankle exoskeletons that assisted plantar flexion. An inverse dynamics analysis, combined with in vivo ultrasound imaging of soleus fascicles and surface electromyography, was used to determine muscle-tendon mechanics and activations. Whole body net metabolic power was obtained from indirect calorimetry. When hopping with spring-loaded exoskeletons, soleus activation was reduced (30-70%) and so was the magnitude of soleus force (peak force reduced by 30%) and the average rate of soleus force generation (by 50%). Although forces were lower, average positive fascicle power remained unchanged, owing to increased fascicle excursion (+4-5 mm). Net metabolic power was reduced with exoskeleton assistance (19%). These findings highlighted that parallel assistance to a muscle with appreciable series elasticity may have some negative consequences, and that the metabolic cost associated with generating force may be more pronounced than the cost of doing work for these muscles.

Timing matters: tuning the mechanics of a muscle-tendon unit by adjusting stimulation phase during cyclic contractions.

ABSTRACT A growing body of research on the mechanics and energetics of terrestrial locomotion has demonstrated that elastic elements acting in series with contracting muscle are critical components of sustained, stable and efficient gait. Far fewer studies have examined how the nervous system modulates muscle–tendon interaction dynamics to optimize ‘tuning’ or meet varying locomotor demands. To explore the fundamental neuromechanical rules that govern the interactions between series elastic elements (SEEs) and contractile elements (CEs) within a compliant muscle–tendon unit (MTU), we used a novel work loop approach that included implanted sonomicrometry crystals along muscle fascicles. This enabled us to decouple CE and SEE length trajectories when cyclic strain patterns were applied to an isolated plantaris MTU from the bullfrog (Lithobates catesbeianus). Using this approach, we demonstrate that the onset timing of muscle stimulation (i.e. stimulation phase) that involves a symmetrical MTU stretch–shorten cycle during active force production results in net zero mechanical power output, and maximal decoupling of CE and MTU length trajectories. We found it difficult to ‘tune’ the muscle–tendon system for strut-like isometric force production by adjusting stimulation phase only, as the zero power output condition involved significant positive and negative mechanical work by the CE. A simple neural mechanism – adjusting muscle stimulation phase – could shift an MTU from performing net zero to net positive (energy producing) or net negative (energy absorbing) mechanical work under conditions of changing locomotor demand. Finally, we show that modifications to the classical work loop paradigm better represent in vivo muscle–tendon function during locomotion. Summary: During cyclic contractions of a muscle–tendon unit, the timing of muscle stimulation onset determines whether the muscle acts like a force producing strut, energy generating motor or energy absorbing brake.

More is not always better: modeling the effects of elastic exoskeleton compliance on underlying ankle muscle–tendon dynamics

Development of robotic exoskeletons to assist/enhance human locomotor performance involves lengthy prototyping, testing, and analysis. This process is further convoluted by variability in limb/body morphology and preferred gait patterns between individuals. In an attempt to expedite this process, and establish a physiological basis for actuator prescription, we developed a simple, predictive model of human neuromechanical adaptation to a passive elastic exoskeleton applied at the ankle joint during a functional task. We modeled the human triceps surae-Achilles tendon muscle tendon unit (MTU) as a single Hill-type muscle, or contractile element (CE), and series tendon, or series elastic element (SEE). This modeled system was placed under gravitational load and underwent cyclic stimulation at a regular frequency (i.e. hopping) with and without exoskeleton (Exo) assistance. We explored the effect that both Exo stiffness (kExo) and muscle activation (Astim) had on combined MTU and Exo (MTU + Exo), MTU, and CE/SEE mechanics and energetics. Model accuracy was verified via qualitative and quantitative comparisons between modeled and prior experimental outcomes. We demonstrated that reduced Astim can be traded for increased kExo to maintain consistent MTU + Exo mechanics (i.e. average positive power (P⁺mech) output) from an unassisted condition (i.e. kExo = 0 kN · m⁻¹). For these regions of parameter space, our model predicted a reduction in MTU force, SEE energy cycling, and metabolic rate (Pmet), as well as constant CE P⁺mech output compared to unassisted conditions. This agreed with previous experimental observations, demonstrating our models predictive ability. Model predictions also provided insight into mechanisms of metabolic cost minimization, and/or enhanced mechanical performance, and we concluded that both of these outcomes cannot be achieved simultaneously, and that one must come at the detriment of the other in a spring-assisted compliant MTU.

Exploiting elasticity: Modeling the influence of neural control on mechanics and energetics of ankle muscle–tendons during human hopping

We present a simplified Hill-type model of the human triceps surae-Achilles tendon complex working on a gravitational-inertial load during cyclic contractions (i.e. vertical hopping). Our goal was to determine the role that neural control plays in governing muscle, or contractile element (CE), and tendon, or series elastic element (SEE), mechanics and energetics within a compliant muscle-tendon unit (MTU). We constructed a 2D parameter space consisting of many combinations of stimulation frequency and magnitude (i.e. neural control strategies). We compared the performance of each control strategy by evaluating peak force and average positive mechanical power output for the system (MTU) and its respective components (CE, SEE), force-length (F-L) and -velocity (F-V) operating point of the CE during active force production, average metabolic rate for the CE, and both MTU and CE apparent efficiency. Our results suggest that frequency of stimulation plays a primary role in governing whole-MTU mechanics. These include the phasing of both activation and peak force relative to minimum MTU length, average positive power, and apparent efficiency. Stimulation amplitude was primarily responsible for governing average metabolic rate and within MTU mechanics, including peak force generation and elastic energy storage and return in the SEE. Frequency and amplitude of stimulation both played integral roles in determining CE F-L operating point, with both higher frequency and amplitude generally corresponding to lower CE strains, reduced injury risk, and elimination of the need for passive force generation in the CE parallel elastic element (PEE).

A Rodent Paw Tracker Using Support Vector Machine

The observation of both human and animal kinematics has proven to be of tremendous scientific value for both understanding biological phenomena and development of medical treatments. Rodents, in particular, are widely used as a model for human disease. Unfortunately, they are also notoriously difficult to use in experiments requiring the use of motion capture (MoCap) for kinematic analyses. Current commercially available technology for rodent MoCap (e.g. Digigait [1, 2], Motorater, Noldus Catwalk [3, 4]) requires either frequent anaesthesia for shaving and tattooing of anatomical landmarks , or retroreflective markers known to agitate rodent subjects. Less automated methods require the use of high speed video capture and manual tracking, which both time and data intensive.

Unconstrained muscle-tendon workloops indicate resonance tuning as a mechanism for elastic limb behavior during terrestrial locomotion.

Significance The fields of terrestrial biomechanics and bio-inspired robotics have identified spring-like limb mechanics as critical to stable and efficient gait. In biological systems, distal muscle groups cycling large amounts of energy in series tendons are a primary source of compliance. To investigate the origins of this behavior, we coupled a biological muscle-tendon to a feedback controlled servomotor simulating the inertial/gravitational environment of terrestrial gait. We drove this bio-robotic system via direct nerve stimulation across a range of frequencies to explore the influence of neural control on muscle-tendon interactions. This study concluded that by matching stimulation frequency to that of the passive biomechanical system, muscle-tendon interactions resulting in spring-like behavior occur naturally and do not require closed-loop neural control. In terrestrial locomotion, there is a missing link between observed spring-like limb mechanics and the physiological systems driving their emergence. Previous modeling and experimental studies of bouncing gait (e.g., walking, running, hopping) identified muscle-tendon interactions that cycle large amounts of energy in series tendon as a source of elastic limb behavior. The neural, biomechanical, and environmental origins of these tuned mechanics, however, have remained elusive. To examine the dynamic interplay between these factors, we developed an experimental platform comprised of a feedback-controlled servo-motor coupled to a biological muscle-tendon. Our novel motor controller mimicked in vivo inertial/gravitational loading experienced by muscles during terrestrial locomotion, and rhythmic patterns of muscle activation were applied via stimulation of intact nerve. This approach was based on classical workloop studies, but avoided predetermined patterns of muscle strain and activation—constraints not imposed during real-world locomotion. Our unconstrained approach to position control allowed observation of emergent muscle-tendon mechanics resulting from dynamic interaction of neural control, active muscle, and system material/inertial properties. This study demonstrated that, despite the complex nonlinear nature of musculotendon systems, cyclic muscle contractions at the passive natural frequency of the underlying biomechanical system yielded maximal forces and fractions of mechanical work recovered from previously stored elastic energy in series-compliant tissues. By matching movement frequency to the natural frequency of the passive biomechanical system (i.e., resonance tuning), muscle-tendon interactions resulting in spring-like behavior emerged naturally, without closed-loop neural control. This conceptual framework may explain the basis for elastic limb behavior during terrestrial locomotion.

Influence of parallel spring-loaded exoskeleton on ankle muscle-tendon dynamics during simulated human hopping

Robotic assistance for rehabilitation and enhancement of human locomotion has become a major goal of biomedical engineers in recent years. While significant progress to this end has been made in the fields of neural interfacing and control systems, little has been done to examine the effects of mechanical assistance on the biomechanics of underlying muscle-tendon systems. Here, we model the effects of mechanical assistance via a passive spring acting in parallel with the triceps surae-Achilles tendon complex during cyclic hopping in humans. We examine system dynamics over a range of biological muscle activation and exoskeleton spring stiffness. We find that, in most cases, uniform cyclic mechanical power production of the coupled system is achieved. Furthermore, unassisted power production can be reproduced throughout parameter space by trading off decreases in muscle activation with increases in ankle exoskeleton spring stiffness. In addition, we show that as mechanical assistance increases the biological muscle-tendon unit becomes less ‘tuned’ resulting in higher mechanical power output from active components of muscle despite large reductions in required force output.

A novel automatic method to track the body and paws of running mice in high speed video

Examining locomotion has improved our basic understanding of motor control and aided in treating motor impairment. Mice are a premier model of human disease and increasingly the model system of choice for basic neuroscience. High frame rates (higher than 150 Hz) are needed to quantify the kinematics of running mice, due to their high stride frequency (up to 10 Hz). Thus manual tracking, especially for multiple markers, becomes time-consuming and impossible for large sample sizes. Therefore, an automated method is necessary. Several methods have been used to automatically or semi-automatically track mice, including commercially available systems (Digigait [1, 2], Motorater, Noldus Catwalk [3, 4]). These systems can be typically prohibitively expensive, and may only provide information about paws during the stand phase. In research and industry approaches to tracking mice have frequently relied on shaving fur and then drawing markers on the skin for subsequent tracking raw video [5], or on the attachment of retroreflective markers, and the use of optical motion capture systems. These methods have the drawback of requiring anesthesia and multiple handlings applications of markers, and the problem of animal removing the attached markers. Here, we develop methods to track and label the body and paws directly from high speed color video, without shaving fur or attaching markers. We analyse data from the C57BL/6 mouse, because it is the most widely used strain in basic research and biomedicine.

Matlab software for impedance spectroscopy designed for neuroscience applications

BACKGROUND Metal electrodes are a mainstay of neuroscience. Characterization of the electrical impedance properties of these cuffs is important to ensure successful and repeatable fabrication, achieve a target impedance, revise novel designs, and quantify the success or failure of implantation and any potential subsequent damage or encapsulation by scar tissue. NEW METHODS Impedances are frequently characterized using lumped-parameter circuit models of the electrode-electrolyte interface. Open-source tools to gather and analyze these frequency sweep data are lacking. Here, we present such software, in the form of Matlab code, which includes a GUI. It automatically acquires frequency sweep data and subsequently fits a simplified Randles model to these data, over a user specified frequency range, providing the user with the model parameter estimates. Also, it can measure an unknown impedance of an element over a range of frequencies, as long as an external resistor can be added for the measurements. RESULTS The tool was tested on five bright platinum nerve cuffs in vitro. The average charge transfer resistance, solution resistance, CPE value, and impedance magnitude were estimated. COMPARISON TO EXISTING METHODS The measured values of the impedance of cuffs were in agreement with the literature (Wei and Grill, 2009). Variation between cuffs fabricated as consistently as possible amounted to 10% for impedance magnitude and 4° for impedance phase. CONCLUSION The results show that this low-cost tool can be used to characterize a cuff across different conditions including after implantation. The latter makes it useful for a longer-term study of electrode viability.

Uncovering the structure of the mouse gait controller: Mice respond to substrate perturbations with adaptations in gait on a continuum between trot and bound

Animals, including humans, have been shown to maintain a gait during locomotion that minimizes the risk of injury and energetic cost. Despite the importance of understanding the mechanisms of gait regulation, ethical and experimental challenges have prevented full exploration of these. Here we present data on the gait response of mice to rapid, precisely timed, spatially confined mechanical perturbations. Our data elucidate that after the mechanical perturbation, the mouse gait response is anisotropic, preferring deviations away from the trot towards bounding, over those towards other gaits, such as walk or pace. We quantified this shift by projecting the observed gait onto the line between trot and bound, in the space of quadrupedal gaits. We call this projection λ. For λ=0, the gait is the ideal trot; for λ=±π, it is the ideal bound. We found that the substrate perturbation caused a significant shift in λ towards bound during the stride in which the perturbation occurred and the following stride (linear mixed effects model: Δλ=0.26±0.07 and Δλ=0.21±0.07, respectively; random effect for animal, p < 0.05 for both strides, n = 8 mice). We hypothesize that this is because the bounding gait is better suited to rapid acceleration or deceleration, and an exploratory analysis of jerk showed that it was significantly correlated with λ (p < 0.05). Understanding how gait is controlled under perturbations can aid in diagnosing gait pathologies and in the design of more agile robots.