Modular hip exoskeleton improves walking function and reduces sedentary time in community-dwelling older adults

Numerous studies showed that postural balance improves through light touch on a stable surface highlighting the importance of haptic information, seemingly downplaying the mechanical contributions of the support. The present study examined the mechanical effects of canes for assisting balance in healthy individuals challenged by standing on a beam. Sixteen participants supported themselves with two canes, one in each hand, and applied minimal, preferred, or maximum force onto the canes. They positioned the canes in the frontal plane or in a tripod configuration. Statistical analysis used a linear mixed model to evaluate the effects on the center of pressure and the center of mass. The canes significantly reduced the variability of the center of pressure and the center of mass to the same level as when standing on the ground. Increasing the exerted force beyond the preferred level yielded no further benefits, although in the preferred force condition, participants exploited the altered mechanics by resting their arms on the canes. The tripod configuration allowed for larger variability of the center of pressure in the task-irrelevant anterior–posterior dimension. High forces had a destabilizing effect on the canes: the displacement of the hand on the cane handle increased with the force. Given this static instability, these results show that using canes can provide not only mechanical benefits but also challenges. From a control perspective, effort can be reduced by resting the arms on the canes and by channeling noise in the task-irrelevant dimensions. However, larger forces exerted onto the canes can also have destabilizing effects and the instability of the canes needs to be counteracted, possibly by arm and shoulder stiffness. Insights into the variety of mechanical effects is important for the design of canes and the instructions of how to use them.     

Reduced reciprocal inhibition during assisted stepping in human spinal cord injury

The aim of this study was to establish the modulation pattern of the reciprocal inhibition exerted from tibialis anterior (TA) group I afferents onto soleus motoneurons during body weight support (BWS) assisted stepping in people with spinal cord injury (SCI). During assisted stepping, the soleus H-reflex was conditioned by percutaneous stimulation of the ipsilateral common peroneal nerve at one fold TA M-wave motor threshold with a single pulse delivered at a short conditioning-test interval. To counteract movement of recording and stimulating electrodes, a supramaximal stimulus at 80–100 ms after the test H-reflex was delivered. Stimuli were randomly dispersed across the step cycle which was divided into 16 equal bins. The conditioned soleus H-reflex was significantly facilitated throughout the stance phase, while during swing no significant changes on the conditioned H-reflex were observed when compared to the unconditioned soleus H-reflex recorded during stepping. Spontaneous clonic activity in triceps surae muscle occurred in multiple phases of the step cycle at a mean frequency of 7 Hz for steps with and without stimulation. This suggests that electrical excitation of TA and soleus group Ia afferents did not contribute to manifestation of ankle clonus. Absent reciprocal inhibition is likely responsible for lack of soleus H-reflex depression in swing phase observed in these patients. The pronounced reduced reciprocal inhibition in stance phase may contribute to impaired levels of co-contraction of antagonistic ankle muscles. Based on these findings, we suggest that rehabilitation should selectively target to transform reciprocal facilitation to inhibition through computer controlled reflex conditioning protocols.     

Primary motor and premotor cortex in implicit sequence learning - evidence for competition between implicit and explicit human motor memory systems

Implicit and explicit memory systems for motor skills compete with each other during and after motor practice. Primary motor cortex (M1) is known to be engaged during implicit motor learning, while dorsal premotor cortex (PMd) is critical for explicit learning. To elucidate the neural substrates underlying the interaction between implicit and explicit memory systems, adults underwent a randomized crossover experiment of anodal transcranial direct current stimulation (AtDCS) applied over M1, PMd or sham stimulation during implicit motor sequence (serial reaction time task, SRTT) practice. We hypothesized that M1‐AtDCS during practice will enhance online performance and offline learning of the implicit motor sequence. In contrast, we also hypothesized that PMd‐AtDCS will attenuate performance and retention of the implicit motor sequence. Implicit sequence performance was assessed at baseline, at the end of acquisition (EoA), and 24 h after practice (retention test, RET). M1‐AtDCS during practice significantly improved practice performance and supported offline stabilization compared with Sham tDCS. Performance change from EoA to RET revealed that PMd‐AtDCS during practice attenuated offline stabilization compared with M1‐AtDCS and sham stimulation. The results support the role of M1 in implementing online performance gains and offline stabilization for implicit motor sequence learning. In contrast, enhancing the activity within explicit motor memory network nodes such as the PMd during practice may be detrimental to offline stabilization of the learned implicit motor sequence. These results support the notion of competition between implicit and explicit motor memory systems and identify underlying neural substrates that are engaged in this competition.     

Appraisals of robotic locomotor exoskeletons for gait: focus group insights from potential users with spinal cord injuries

Abstract

Purpose: To describe appraisals of robotic exoskeletons for locomotion by potential users with spinal cord injuries, their perceptions of device benefits and limitations, and recommendations for manufacturers and therapists regarding device use.

Materials and methods: We conducted focus groups at three regional rehabilitation hospitals and used thematic analysis to define themes.

Results: Across four focus groups, 35 adults participated; they were predominantly middle-aged, male, and diverse in terms of race and ethnicity, well educated, and not working. Participants had been living with SCI an average of two decades. Most participants were aware of exoskeletons. Some were enthusiastic about the usability of the devices while others were more circumspect. They had many questions about device affordability and usability, and were discerning in their appraisal of benefits and suitability to their particular circumstances. They reflected on device cost, the need for caregiver assistance, use of hands, and environmental considerations. They weighed the functional benefits relative to the cost of preferred activities. Their recommendations focused on cost, battery life, and independent use.

Conclusions: Potential users’ appraisals of mobility technology reflect a nuanced appreciation of device costs; functional, social, and psychological benefits; and limitations. Results provide guidance to therapists and manufacturers regarding device use.

• Implications for Rehabilitation

• Potential users of robotic locomotor exoskeletons with spinal cord injuries appreciate the functional, social, and psychological benefits that these devices may offer.

• Their appraisals reflect nuanced consideration of device cost and features, and the suitability of the assistive technology to their circumstances.

• They recommend that manufacturers focus on reducing cost, extending battery life, and features that allow independent use.

     

Postural and Metabolic Benefits of Using a Forearm Support Walker in Older Adults With Impairments

Objective

To investigate the postural and metabolic benefits a walker with adjustable elbow support (LifeWalker [LW]) can provide for ambulation in population with impairment. The clinical outcomes from the elbow support walker will be compared with standard rollator (SR) and participants predicate device (PD).

Corticospinal excitability during walking in humans with absent and partial body weight support

Objective

To establish changes in corticospinal excitability with absent and partial body weight support (BWS), and determine test–retest reliability of motor evoked potentials (MEPs) recordings during stepping in healthy humans.

Methods

The tibialis anterior (TA) and soleus MEPs during stepping at 0 and at 25 BWS were recorded in two experimental sessions in the same subjects. Transcranial magnetic stimulation was delivered randomly across the step cycle at 1.2 × TA MEP resting threshold. The non-stimulated associated electromyogram (EMG) was subtracted from the TA and soleus MEPs at identical time windows and bins of the step cycle, and the resultant values were normalized to the maximal homologous EMG activity during stepping. The relationship between MEPs and background EMG activity was determined for each BWS level and session tested.

Results

The TA MEPs were facilitated at heel contact, progressively decreased during the stance phase, and facilitated throughout the swing phase of the step cycle. In contrast, the soleus MEPs were progressively increased at early-stance, depressed at the stance-to-swing transition, and remained depressed throughout the swing phase. The TA and soleus MEPs were modulated in a similar pattern across sessions at 0 and at 25 BWS, and were linearly related to the associated background EMG activity.

Conclusions

These results provide evidence that reduced body weight loading does not alter the strength of corticospinal excitability, and that MEPs can be reliably recorded at different sessions during stepping in healthy humans.

Significance

A rehabilitation strategy to restore gait in neurological disorders utilizes BWS during stepping on a motorized treadmill. Based on our findings, the strength of corticospinal drive will not be affected negatively during stepping under conditions of partial body loading.     

Modulation of reciprocal and presynaptic inhibition during robotic-assisted stepping in humans

ObjectiveTo establish the modulation pattern of reciprocal inhibition and presynaptic inhibition of soleus Ia afferents during robot-assisted stepping in healthy subjects.MethodsDuring stepping, the soleus H-reflex was conditioned by percutaneous stimulation of the ipsilateral common peroneal nerve with a single pulse at stimulation intensities that ranged from 0.9 to 1.2 TA M-wave motor thresholds across subjects. To control for movement of recording and stimulating electrodes, a supramaximal stimulus 80 ms after the conditioned and/or unconditioned H-reflexes was delivered to the posterior tibial nerve. The short (2, 3, 4 ms) and long (60–80 ms) conditioning-test intervals at which the largest amount of reflex depression was observed with the subjects seated were utilized during stepping. Stimuli were randomly dispersed across the step cycle which was divided into 16 equal bins.ResultsReciprocal inhibition exerted from flexor group I afferents onto soleus motoneurons was decreased at mid-stance and increased and late-stance and throughout the swing phase. Presynaptic inhibition of soleus Ia afferents was increased at heel strike and decreased at late-stance and early swing phases.ConclusionReciprocal inhibition between ankle antagonistic muscles and presynaptic inhibition of soleus Ia afferents are modulated in a similar pattern to that reported during walking on a treadmill with full weight bearing and without robot-assisted leg movement.SignificanceThe activity of spinal interneuronal circuits engaged in patterned locomotor activity supports a reciprocal gait pattern during robot-assisted stepping in healthy humans.     

Automatic analysis of EMG during clonus

Clonus can disrupt daily activities after spinal cord injury. Here an algorithm was developed to automatically detect contractions during clonus in 24 h electromyographic (EMG) records. Filters were created by non-linearly scaling a Mother (Morlet) wavelet to envelope the EMG using different frequency bands. The envelope for the intermediate band followed the EMG best (74.8-193.9 Hz). Threshold and time constraints were used to reduce the envelope peaks to one per contraction. Energy in the EMG was measured 50 ms either side of each envelope (contraction) peak. Energy values at 5% and 95% maximal defined EMG start and end time, respectively. The algorithm was as good as a person at identifying contractions during clonus (p = 0.946, n = 31 spasms, 7 subjects with cervical spinal cord injury), and marking start and end times to determine clonus frequency (intra class correlation coefficient, α: 0.949), contraction intensity using root mean square EMG (α: 0.997) and EMG duration (α: 0.852). On average the algorithm was 574 times faster than manual analysis performed independently by two people (p ≤ 0.001). This algorithm is an important tool for characterization of clonus in long-term EMG records.     

Automated,multiparametric monitoring of respiratory biomarkers and vital signs in clinical and home settings for COVID-19 patients

Capabilities in continuous monitoring of key physiological parameters of disease have never been more important than in the context of the global COVID-19 pandemic. Soft, skin-mounted electronics that incorporate high-bandwidth, miniaturized motion sensors enable digital, wireless measurements of mechanoacoustic (MA) signatures of both core vital signs (heart rate, respiratory rate, and temperature) and underexplored biomarkers (coughing count) with high fidelity and immunity to ambient noises. This paper summarizes an effort that integrates such MA sensors with a cloud data infrastructure and a set of analytics approaches based on digital filtering and convolutional neural networks for monitoring of COVID-19 infections in sick and healthy individuals in the hospital and the home. Unique features are in quantitative measurements of coughing and other vocal events, as indicators of both disease and infectiousness. Systematic imaging studies demonstrate correlations between the time and intensity of coughing, speaking, and laughing and the total droplet production, as an approximate indicator of the probability for disease spread. The sensors, deployed on COVID-19 patients along with healthy controls in both inpatient and home settings, record coughing frequency and intensity continuously, along with a collection of other biometrics. The results indicate a decaying trend of coughing frequency and intensity through the course of disease recovery, but with wide variations across patient populations. The methodology creates opportunities to study patterns in biometrics across individuals and among different demographic groups.

As of December 26, The Centers for Disease Control and Prevention (CDC) tabulations indicate over 18 million recorded cases of COVID-19 and more than 329,592 in deaths in the United States (1). Accurate and widespread testing is a key component of the response to this pandemic (2). Although the capacity and availability of COVID-19 molecular diagnostics continues to increase, shortcomings follow from variabilities in the accuracy of the tests, constraints in materials and supplies, long turnaround times associated with certain tests, inadequate access to testing sites, and a lack of human resources (3). An additional challenge is in limited prognostic tools to assess the trajectory of infection and the eventual need for hospitalization or mechanical ventilation. The CDC confirms that COVID-19 can be contracted via airborne transmission along with contact and droplet transmission—features that underscore the need to improve capabilities in risk stratification of exposures via contact tracing and to ensure sufficient quarantining for recovering individuals.To address some of these needs, a range of digital health tools, from mobile applications for collecting self-reported symptoms to consumer wearable devices and clinical-grade medical sensors for tracking physiological status, are under development and in initial stages of deployment (4). Researchers at FitBit report the ability to identify infection with COVID-19 via four previous days of data collected from their wrist-worn devices to yield overnight heart rate, respiratory rate, and heart rate variability (5). Others claim similar detection capabilities with alternative wrist-based devices (6). Several ongoing large-scale trials aim to evaluate these wearables for early detection of COVID-19 infection, from smart rings (Oura Ring) to skin-interfaced patches [VitalConnect (7), Philips (8), Sonica (9)], to other smart watches [e.g., Empatica (10)] with support from various federal agencies. Devices that mount on the finger or wrist can monitor some subset of conventional vital signs (11–15), such as heart rate. Loose interfaces at these body locations, however, limit the range of detectable physiological activities, particularly respiratory signals (16, 17). The inability to capture complex health information reduces the potential for precise and reliable analysis (18). Development of robust metrics for early detection and disease tracking requires multiparametric operation across different digital biomarkers and unconventional metrics relevant to the disease of interest. Challenges remain in addressing these requirements simultaneously while maintaining simplicity and ease of use of the sensing system, as is necessary for practical deployment at scale in remote, continuous monitoring settings (19).As COVID-19 is a respiratory disease, cough and other sounds from the thoracic cavity, trachea, and esophagus are examples of highly relevant biometrics. Laboratory-scale studies demonstrate cough-based diagnoses of diverse respiratory diseases through measurements of frequency (20), intensity (21), persistency (22), and unique audio features (23). Investigations on audio recording data show differences between COVID-19 positive and negative subjects’ vocalizing patterns including phonation of speech (24, 25), breathing, and coughing sounds (26–29). The results may suggest possibilities for disease monitoring in asymptomatic patients. Recent work applies voice profiling and computer audition to track cough, speech, respiratory, and other sounds for risk assessment and diagnosis of COVID-19 (30, 31). Monitoring cough and other vocal events (speaking, laughing, etc.) not only provides a signature of disease but also has potential in generating metrics of infectiousness, as these mechanisms yield aerosols/droplets that contribute to virus transmission (32–34). Previous studies show that the total volume of aerosols correlate with the loudness and duration of vocal events. Measurements of the timing and intensity of sounds may, therefore, serve as reliable means to quantify one aspect associated with risks of spreading the disease (35).Point-of-care or semicontinuous methods for quantifying coughing or other vocal activities rely on electromyography, respiratory inductive plethysmography, accelerometry, or auditory recordings captured with one or several sensors, sometimes with other exploratory approaches (e.g., the nasal thermistor or the electrocardiography) (36–41). Digital signal processing followed by machine learning algorithms often serves as the basis for classification (42–53). Microphone-based methods prevail due to their widespread availability and their alignment with large crowd-sourced datasets (e.g., COUGHVID, HealthMode, DetectNow, VoiceMed). A key challenge is that background sounds and/or environmental noises frustrate robust and accurate measurements. Measurements of loudness can be unreliable because they depend on the separation between the device and the subject. Most importantly, audio recordings raise privacy and legal issues, thereby limiting the scale of application.The results presented here bypass these disadvantages, to allow continuous assessments of respiratory biomarkers correlative to health status and droplet/aerosol production, with additional information on a range of traditional vital signs. Here, a simple, wireless monitoring device (54) combines with a cloud interface and a data analytics approach to allow continuous monitoring of a breadth of conventional (e.g., heart rate, respiratory rate, physical activity, body orientation, and temperature) and unconventional (e.g., coughing, speaking) physiological parameters of direct relevance to COVID-19. The results serve as a quantitative basis for 1) detecting early signs of symptoms in health care workers and other high-risk populations, 2) monitoring symptomatic progression of infected individuals, and 3) tracking responses to therapeutics in clinical settings. In addition, systematic studies presented here indicate that coughing, speaking, and laughing events measured with these devices correlate to the total amount of droplet production. This link offers an opportunity to quantify the infectiousness of individuals, as critical information in caring for patients and for improved risk stratification in the context of contact tracing and individual quarantines.Pilot studies on COVID-19 patients at an academic medical center (Northwestern Memorial Hospital) and a rehabilitation hospital (Shirley Ryan AbilityLab) include 3,111 h of data spanning a total of 363 d from 37 patients (20 females, 17 males), in an overall implementation that supports automated operation, with minimal user burden. Long-term monitoring reveals trends in various parameters, including coughing frequency, following the test-positive date for eight patients (four females, four males) over more than 7 d. Evaluations across 27 patients (15 females, 12 males) with ages between 21 and 75 y reveal diverse coughing patterns across individuals and consistent trends during the recovery process.