Brain-machine interface based on deep learning to control asynchronously a lower-limb robotic exoskeleton: a case-of-study.

BACKGROUND: This research focused on the development of a motor imagery (MI) based brain-machine interface (BMI) using deep learning algorithms to control a lower-limb robotic exoskeleton. The study aimed to overcome the limitations of traditional BMI approaches by leveraging the advantages of deep learning, such as automated feature extraction and transfer learning. The experimental protocol to evaluate the BMI was designed as asynchronous, allowing subjects to perform mental tasks at their own will. METHODS: A total of five healthy able-bodied subjects were enrolled in this study to participate in a series of experimental sessions. The brain signals from two of these sessions were used to develop a generic deep learning model through transfer learning. Subsequently, this model was fine-tuned during the remaining sessions and subjected to evaluation. Three distinct deep learning approaches were compared: one that did not undergo fine-tuning, another that fine-tuned all layers of the model, and a third one that fine-tuned only the last three layers. The evaluation phase involved the exclusive closed-loop control of the exoskeleton device by the participants' neural activity using the second deep learning approach for the decoding. RESULTS: The three deep learning approaches were assessed in comparison to an approach based on spatial features that was trained for each subject and experimental session, demonstrating their superior performance. Interestingly, the deep learning approach without fine-tuning achieved comparable performance to the features-based approach, indicating that a generic model trained on data from different individuals and previous sessions can yield similar efficacy. Among the three deep learning approaches compared, fine-tuning all layer weights demonstrated the highest performance. CONCLUSION: This research represents an initial stride toward future calibration-free methods. Despite the efforts to diminish calibration time by leveraging data from other subjects, complete elimination proved unattainable. The study's discoveries hold notable significance for advancing calibration-free approaches, offering the promise of minimizing the need for training trials. Furthermore, the experimental evaluation protocol employed in this study aimed to replicate real-life scenarios, granting participants a higher degree of autonomy in decision-making regarding actions such as walking or stopping gait.

Effects of ankle exoskeleton assistance and plantar pressure biofeedback on incline walking mechanics and muscle activity in cerebral palsy.

Ankle dysfunction affects more than 50 % of people with cerebral palsy, resulting in atypical gait patterns that impede lifelong mobility. Incline walking requires increased lower limb effort and is a promising intervention that targets lower-limb extensor muscles. A concern when prescribing incline walking to people with gait deficits is that this exercise may be too challenging or reinforce unfavorable gait patterns. This study aims to investigate how ankle exoskeleton assistance and plantar pressure biofeedback would affect gait mechanics and muscle activity during incline walking in CP. We recruited twelve children and young adults with CP. Participants walked with ankle assistance alone, biofeedback alone, and the combination while we assessed ankle, knee, and hip mechanics, and plantar flexor and knee extensor activity. Compared to incline walking without assistance or biofeedback, ankle assistance alone reduced the peak biological ankle moment by 12 % (p < 0.001) and peak soleus activity by 8 % (p = 0.013); biofeedback alone increased the biological ankle moment (4 %, p = 0.037) and power (19 %, p = 0.012), and plantar flexor activities by 9 - 27 % (p ≤ 0.026); assistance-plus-biofeedback increased biological ankle and knee power by 34 % and 17 %, respectively (p ≤ 0.05). The results indicate that both ankle exoskeleton assistance and plantar pressure biofeedback can effectively modify lower limb mechanics and muscular effort during incline walking in CP. These techniques may help in establishing personalized gait training interventions by providing the ability to adjust intensity and biomechanical focus over time.

[Application of exoskeleton with functional electrostimulation for rebalancing of patients in acute and early recovery periods of ischemic stroke]. / Primenenie ekzoskeleta s funktsional'noi elektrostimulyatsiei v vosstanovlenii ravnovesiya u patsientov v ostrom i rannem vosstanovitel'nom periodakh ishemicheskogo insul'ta.

Static motor disorders are the main cause of falls and decrease of daily activity in patients after previous ischemic stroke. OBJECTIVE: To study impact of robotic mechanotherapy with functional electrostimulation (FES) on rebalancing in patients in acute and early recovery periods of ischemic stroke. MATERIAL AND METHODS: The number of patients equal 60, divided into 2 groups, were examined. Study group included 30 patients, who were treated in an exoskeleton with FES; control group consisted of 30 patients, whose walking and balance recovering was carried out with exercise therapy. RESULTS: The study group showed a more pronounced improvement of balance indicators by Tinetti scale compared to the control group (11 [8; 13] against 8 [6; 12]; p=0.0281) at the end of treatment. Patients of study group demonstrated significantly better results both by Tinetti scale and stabilometrical parameters in acute period of stroke compared to the control group. CONCLUSION: Application of exoskeleton with FES in patients in acute and early periods of ischemic stroke contributes (to a better balance recovery), which may be achieved by long-lasting verticalization and large repetition of the correct walking pattern during rehabilitative trainings.

BCI-Based Control for Ankle Exoskeleton T-FLEX: Comparison of Visual and Haptic Stimuli with Stroke Survivors.

Brain-computer interface (BCI) remains an emerging tool that seeks to improve the patient interaction with the therapeutic mechanisms and to generate neuroplasticity progressively through neuromotor abilities. Motor imagery (MI) analysis is the most used paradigm based on the motor cortex's electrical activity to detect movement intention. It has been shown that motor imagery mental practice with movement-associated stimuli may offer an effective strategy to facilitate motor recovery in brain injury patients. In this sense, this study aims to present the BCI associated with visual and haptic stimuli to facilitate MI generation and control the T-FLEX ankle exoskeleton. To achieve this, five post-stroke patients (55-63 years) were subjected to three different strategies using T-FLEX: stationary therapy (ST) without motor imagination, motor imagination with visual stimulation (MIV), and motor imagination with visual-haptic inducement (MIVH). The quantitative characterization of both BCI stimuli strategies was made through the motor imagery accuracy rate, the electroencephalographic (EEG) analysis during the MI active periods, the statistical analysis, and a subjective patient's perception. The preliminary results demonstrated the viability of the BCI-controlled ankle exoskeleton system with the beta rebound, in terms of patient's performance during MI active periods and satisfaction outcomes. Accuracy differences employing haptic stimulus were detected with an average of 68% compared with the 50.7% over only visual stimulus. However, the power spectral density (PSD) did not present changes in prominent activation of the MI band but presented significant variations in terms of laterality. In this way, visual and haptic stimuli improved the subject's MI accuracy but did not generate differential brain activity over the affected hemisphere. Hence, long-term sessions with a more extensive sample and a more robust algorithm should be carried out to evaluate the impact of the proposed system on neuronal and motor evolution after stroke.

A Robotic System with EMG-Triggered Functional Eletrical Stimulation for Restoring Arm Functions in Stroke Survivors.

BACKGROUND: Robotic systems combined with Functional Electrical Stimulation (FES) showed promising results on upper-limb motor recovery after stroke, but adequately-sized randomized controlled trials (RCTs) are still missing. OBJECTIVE: To evaluate whether arm training supported by RETRAINER, a passive exoskeleton integrated with electromyograph-triggered functional electrical stimulation, is superior to advanced conventional therapy (ACT) of equal intensity in the recovery of arm functions, dexterity, strength, activities of daily living, and quality of life after stroke. METHODS: A single-blind RCT recruiting 72 patients was conducted. Patients, randomly allocated to 2 groups, were trained for 9 weeks, 3 times per week: the experimental group performed task-oriented exercises assisted by RETRAINER for 30 minutes plus ACT (60 minutes), whereas the control group performed only ACT (90 minutes). Patients were assessed before, soon after, and 1 month after the end of the intervention. Outcome measures were as follows: Action Research Arm Test (ARAT), Motricity Index, Motor Activity Log, Box and Blocks Test (BBT), Stroke Specific Quality of Life Scale (SSQoL), and Muscle Research Council. RESULTS: All outcomes but SSQoL significantly improved over time in both groups (P < .001); a significant interaction effect in favor of the experimental group was found for ARAT and BBT. ARAT showed a between-group change of 11.5 points (P = .010) at the end of the intervention, which increased to 13.6 points 1 month after. Patients considered RETRAINER moderately usable (System Usability Score of 61.5 ± 22.8). CONCLUSIONS: Hybrid robotic systems, allowing to perform personalized, intensive, and task-oriented training, with an enriched sensory feedback, was superior to ACT in improving arm functions and dexterity after stroke.

Synthesis of a Removable Cytoprotective Exoskeleton by Tea Polyphenol Complexes for Living Cell Encapsulation.

Cell encapsulation is a chemical tool for endowing living cells with exogenous properties and enhancing their in vitro tolerance against lethal factors, which has shown promising prospects and potential applications in many fields such as cell transplantation, drug delivery, and tissue engineering. One-pot precipitation of a polyphenol-metal complex on cells protects cells from UV irradiation and lytic enzymes. However, the involvement of metal ions brings side effects on cell viability and growth. Moreover, an external removal agent is needed for cell division and growth. Herein, a polymer shell composed of hydrogen bonded constituents without affecting cell viability and growth by the precipitation of tea polyphenol and polyvinyl pyrrolidone is reported. The formation of the polymer shell was verified by the Au nanoparticle's laser scanning confocal reflectance and quartz crystal microbalance measurement. The thickness of the shell was managed by the concentration of the complex. When exposed to UV irradiation for 15 or 30 min, polymer-coating-protected Saccharomyces cerevisiae (yeast) had much higher cell viability than the native one. Exposed to a high temperature environment (60 °C), most of the coated yeasts survived in contrast to uncoated ones. For the cell division and growth curve, the polymer coating with various thicknesses had no difference to the native one, which indicated no suppression of cell growth and no external side effects involved. As applied to mammalian HeLa cells under UV irradiation for 15 min, the coated cells had an obvious higher cell viability than that of untreated ones. Therefore, the tea polyphenol-poly(vinylpyrrolidone) shell is a versatile tool for chemically controlling the external properties of cells without side effects on cell viability and growth.

Wearable Biofeedback Improves Human-Robot Compliance during Ankle-Foot Exoskeleton-Assisted Gait Training: A Pre-Post Controlled Study in Healthy Participants.

The adjunctive use of biofeedback systems with exoskeletons may accelerate post-stroke gait rehabilitation. Wearable patient-oriented human-robot interaction-based biofeedback is proposed to improve patient-exoskeleton compliance regarding the interaction torque's direction (joint motion strategy) and magnitude (user participation strategy) through auditory and vibrotactile cues during assisted gait training, respectively. Parallel physiotherapist-oriented strategies are also proposed such that physiotherapists can follow in real-time a patient's motor performance towards effective involvement during training. A preliminary pre-post controlled study was conducted with eight healthy participants to conclude about the biofeedback's efficacy during gait training driven by an ankle-foot exoskeleton and guided by a technical person. For the study group, performance related to the interaction torque's direction increased during (p-value = 0.07) and after (p-value = 0.07) joint motion training. Further, the performance regarding the interaction torque's magnitude significantly increased during (p-value = 0.03) and after (p-value = 68.59×10-3) user participation training. The experimental group and a technical person reported promising usability of the biofeedback and highlighted the importance of the timely cues from physiotherapist-oriented strategies. Less significant improvements in patient-exoskeleton compliance were observed in the control group. The overall findings suggest that the proposed biofeedback was able to improve the participant-exoskeleton compliance by enhancing human-robot interaction; thus, it may be a powerful tool to accelerate post-stroke ankle-foot deformity recovery.

Sensory Integration in Human Movement: A New Brain-Machine Interface Based on Gamma Band and Attention Level for Controlling a Lower-Limb Exoskeleton.

Brain-machine interfaces (BMIs) can improve the control of assistance mobility devices making its use more intuitive and natural. In the case of an exoskeleton, they can also help rehabilitation therapies due to the reinforcement of neuro-plasticity through repetitive motor actions and cognitive engagement of the subject. Therefore, the cognitive implication of the user is a key aspect in BMI applications, and it is important to assure that the mental task correlates with the actual motor action. However, the process of walking is usually an autonomous mental task that requires a minimal conscious effort. Consequently, a brain-machine interface focused on the attention to gait could facilitate sensory integration in individuals with neurological impairment through the analysis of voluntary gait will and its repetitive use. This way the combined use of BMI+exoskeleton turns from assistance to restoration. This paper presents a new brain-machine interface based on the decoding of gamma band activity and attention level during motor imagery mental tasks. This work also shows a case study tested in able-bodied subjects prior to a future clinical study, demonstrating that a BMI based on gamma band and attention-level paradigm allows real-time closed-loop control of a Rex exoskeleton.

Toward a hybrid exoskeleton for crouch gait in children with cerebral palsy: neuromuscular electrical stimulation for improved knee extension.

BACKGROUND: Neuromuscular Electrical Stimulation (NMES) has been utilized for many years in cerebral palsy (CP) with limited success despite its inherent potential for improving muscle size and/or strength, inhibiting or reducing spasticity, and enhancing motor performance during functional activities such as gait. While surface NMES has been shown to successfully improve foot drop in CP and stroke, correction of more complex gait abnormalities in CP such as flexed knee (crouch) gait remains challenging due to the level of stimulation needed for the quadriceps muscles that must be balanced with patient tolerability and the ability to deliver NMES assistance at precise times within a gait cycle. METHODS: This paper outlines the design and evaluation of a custom, noninvasive NMES system that can trigger and adjust electrical stimulation in real-time. Further, this study demonstrates feasibility of one possible application for this digitally-controlled NMES system as a component of a pediatric robotic exoskeleton to provide on-demand stimulation to leg muscles within specific phases of the gait cycle for those with CP and other neurological disorders who still have lower limb sensation and volitional control. A graphical user interface was developed to digitally set stimulation parameters (amplitude, pulse width, and frequency), timing, and intensity during walking. Benchtop testing characterized system delay and power output. System performance was investigated during a single session that consisted of four overground walking conditions in a 15-year-old male with bilateral spastic CP, GMFCS Level III: (1) his current Ankle-Foot Orthosis (AFO); (2) unassisted Exoskeleton; (3) NMES of the vastus lateralis; and (4) NMES of the vastus lateralis and rectus femoris. We hypothesized in this participant with crouch gait that NMES triggered with low latency to knee extensor muscles during stance would have a modest but positive effect on knee extension during stance. RESULTS: The system delivers four channels of NMES with average delays of 16.5 ± 13.5 ms. Walking results show NMES to the vastus lateralis and rectus femoris during stance immediately improved mean peak knee extension during mid-stance (p = 0.003*) and total knee excursion (p = 0.009*) in the more affected leg. The electrical design, microcontroller software and graphical user interface developed here are included as open source material to facilitate additional research into digitally-controlled surface stimulation ( github.com/NIHFAB/NMES ). CONCLUSIONS: The custom, digitally-controlled NMES system can reliably trigger electrical stimulation with low latency. Precisely timed delivery of electrical stimulation to the quadriceps is a promising treatment for crouch. Our ultimate goal is to synchronize NMES with robotic knee extension assistance to create a hybrid NMES-exoskeleton device for gait rehabilitation in children with flexed knee gait from CP as well as from other pediatric disorders. TRIAL REGISTRATION: clinicaltrials.gov, ID: NCT01961557 . Registered 11 October 2013; Last Updated 27 January 2020.

From sporadic single genes to a broader transcriptomic approach: Insights into the formation of the biomineralized exoskeleton in decapod crustaceans.

One fundamental character common to pancrustaceans (Crustacea and Hexapoda) is a mineralized rigid exoskeleton whose principal organic components are chitin and proteins. In contrast to traditional research in the field that has been devoted to the structural and physicochemical aspects of biomineralization, the present study explores transcriptomic aspects of biomineralization as a first step towards adding a complementary molecular layer to this field. The rigidity of the exoskeleton in pancrustaceans dictates essential molt cycles enabling morphological changes and growth. Thus, formation and mineralization of the exoskeleton are concomitant to the timeline of the molt cycle. Skeletal proteinaceous toolkit elements have been discovered in previous studies using innovative molt-related binary gene expression patterns derived from transcriptomic libraries representing the major stages comprising the molt cycle of the decapod crustacean Cherax quadricarinatus. Here, we revisited some prominent exoskeleton-related structural proteins encoding and, using the above molt-related binary pattern methodology, enlarged the transcriptomic database of C. quadricarinatus. The latter was done by establishing a new transcriptomic library of the cuticle forming epithelium and molar tooth at four different molt stages (i.e., inter-molt, early pre-molt, late pre-molt and post-molt) and incorporating it to a previous transcriptome derived from the gastroliths and mandible. The wider multigenic approach facilitated by the newly expanded transcriptomic database not only revisited single genes of the molecular toolkit, but also provided both scattered and specific information that broaden the overview of proteins and gene clusters which are involved in the construction and biomineralization of the exoskeleton in decapod crustaceans.

A Case of Neuromuscular Electrical Stimulation for Childhood Stroke Hyperkinesis: A Brief Report.

AIM: Some conditions within specific populations are so rare rigorous evidence is unavailable. Childhood hyperkinesis is one example, yet presents an opportunity to examine sensation's contribution to motor function. METHODS: The patient experienced functional difficulty from hyperkinesis as a result of childhood stroke. Home-based passive neuromuscular electrical stimulation (NMES) was implemented an hour/day, six days/week, over 6 weeks (36 hours). Clinical and robotic measures (Assisting Hand Assessment, Box and Block Test, Jebsen Taylor Test of Hand Function, Kinarm) were administered before and after the intervention and at 9 months. RESULTS: NMES was feasible and well tolerated. Clinically important gains of arm function were maintained at 9 months. Robotic measures showed improved hyperkinesis, namely reduced movement segmentation and improved target approximation, in addition to improved proprioceptive function after NMES. CONCLUSION: This case study illustrates the use of NMES within a previously unexplored population and highlights the potential importance of sensory systems to motor gains.

Mobilising the patient: With emphasis on innovative technologies.

Treatment of trauma patients and fractures has changed dramatically throughout the years. From conservative methods to nowadays various kinds of screws, pins, plates and nails for optimal fixation of fractures. This lead to changes in post-operative management as well, from bedrest to (partial) weight bearing. Some patients however have very limited to no ability to mobilise, such as critical ill patients on the Intensive Care Unit, amputees or spinal cord injured patients. Due to innovations such as hydrotherapy, osseointegrated prosthesis and exoskeletons, even these people can mobilise. Thanks to innovations like these an increasing number of trauma patients are able to fully reintegrate into community life and get back to an active and independent life style.

Effects of Electromechanical Exoskeleton-Assisted Gait Training on Walking Ability of Stroke Patients: A Randomized Controlled Trial.

OBJECTIVE: To assess the efficacy of electromechanical exoskeleton-assisted gait training on walking ability of stroke patients based on ambulatory function, muscle strength, balance, gait speed, and capacity. DESIGN: Randomized controlled trial. SETTING: University rehabilitation hospital. PARTICIPANTS: Individuals (N=40) with stroke who could stand alone. INTERVENTIONS: Patients were randomly assigned to control and experimental groups. The control group underwent physical therapist-assisted gait training by conventional method. The experimental group underwent electromechanical gait training assisted by an exoskeleton device. Both types of gait training were performed for 30 minutes each day. The therapeutic interventions were provided for 5 days a week for a period of 4 weeks in both groups. MAIN OUTCOME MEASURES: Functional ambulatory category (FAC) before and after gait training. Changes in FAC were the primary outcomes to evaluate the efficacy of electromechanical exoskeleton-assisted gait training. Changes in mobility, walking speed, walking capacity, leg muscle strength, daily activity, and balance were secondary outcomes. RESULTS: FAC in the control group was 2.44±1.55 in the pretraining and 2.75±1.53 in the post-training. FAC in the experimental group was 3.22±1.31 in the pretraining and 3.78±1.44 in the post-training. Although FAC between pre- and post-training sessions improved in both groups, the changes in FAC were statistically significant in the experimental group alone. Most secondary outcomes in both groups also showed improvement after gait training. However, the differential outcomes were not varied between the 2 groups after adjusting the data for age and stroke duration. We did not exclude patients based on time since stroke onset. The average stroke duration was 530.11±389.21 days in the experimental group. The changes in FAC of the experimental group were negatively correlated with stroke duration. No adverse events were noticed during gait training in either group. CONCLUSIONS: Electromechanical exoskeleton-assisted gait training is as effective as conventional gait training by a physical therapist when administered by a gait trainer. As an overground walking system without harness, electromechanical exoskeleton replaced a physical therapist in assisted gait training for patients who stand alone. Because the ambulatory function of stroke patients was affected negatively by stroke duration, the effect of electromechanical-assisted gait training might decline with increased stroke duration.

Consequences of prey exoskeleton content for predator feeding and digestion: black widow predation on larval versus adult mealworm beetles.

Predators often feed on a wide range of prey that can vary in behavior, morphology, and physiology. The net benefits that predators gain from prey are likely related to both prey nutrient content and prey morphology or defenses. For invertebrates, the exoskeleton is a morphological trait that varies widely among species and during ontogeny and could affect nutrient extraction by predators. The goal of this study was to determine how prey exoskeleton content affected predator nutrient intake, assimilation, and excretion by comparing spiders feeding on either larval or adult mealworms of similar size. We found that the proportion of prey energy invested in digestion was greatest in spiders consuming adult mealworm beetles which had higher amounts of exoskeleton than larvae. Further, spiders extracted a greater proportion of elements, macronutrients, and energy from the larval mealworms, which had lower amounts of exoskeleton. Interestingly, total nitrogen content of prey was not a predictor of nitrogen assimilation as spiders assimilated more nitrogen from the larval mealworms, which had lower total nitrogen content. While adult beetles had higher total nitrogen content, their discarded remains of prey had large amounts of nitrogen that was nutritionally unavailable for spiders (i.e., exoskeleton). These results suggest that prey exoskeleton can affect assimilation efficiency by predators, and that a combination of macronutrient and elemental analyses may be needed to examine the quality of prey for predators and the potential consequences of predation for nutrient flows (e.g., consumer assimilation, egestion, and excretion) in ecosystems.

Improving Real-Time Lower Limb Motor Imagery Detection Using tDCS and an Exoskeleton.

The aim of this work was to test if a novel transcranial direct current stimulation (tDCS) montage boosts the accuracy of lower limb motor imagery (MI) detection by using a real-time brain-machine interface (BMI) based on electroencephalographic (EEG) signals. The tDCS montage designed was composed of two anodes and one cathode: one anode over the right cerebrocerebellum, the other over the motor cortex in Cz, and the cathode over FC2 (using the International 10-10 system). The BMI was designed to detect two MI states: relax and gait MI; and was based on finding the power at the frequency which attained the maximum power difference between the two mental states at each selected EEG electrode. Two different single-blind experiments were conducted, E1 and a pilot test E2. E1 was based on visual cues and feedback and E2 was based on auditory cues and a lower limb exoskeleton as feedback. Twelve subjects participated in E1, while four did so in E2. For both experiments, subjects were separated into two equally-sized groups: sham and active tDCS. The active tDCS group achieved 12.6 and 8.2% higher detection accuracy than the sham group in E1 and E2, respectively, reaching 65 and 81.6% mean detection accuracy in each experiment. The limited results suggest that the exoskeleton (E2) enhanced the detection of the MI tasks with respect to the visual feedback (E1), increasing the accuracy obtained in 16.7 and 21.2% for the active tDCS and sham groups, respectively. Thus, the small pilot study E2 indicates that using an exoskeleton in real-time has the potential of improving the rehabilitation process of cerebrovascular accident (CVA) patients, but larger studies are needed in order to further confirm this claim.

EEG-Based Control for Upper and Lower Limb Exoskeletons and Prostheses: A Systematic Review.

Electroencephalography (EEG) signals have great impact on the development of assistive rehabilitation devices. These signals are used as a popular tool to investigate the functions and the behavior of the human motion in recent research. The study of EEG-based control of assistive devices is still in early stages. Although the EEG-based control of assistive devices has attracted a considerable level of attention over the last few years, few studies have been carried out to systematically review these studies, as a means of offering researchers and experts a comprehensive summary of the present, state-of-the-art EEG-based control techniques used for assistive technology. Therefore, this research has three main goals. The first aim is to systematically gather, summarize, evaluate and synthesize information regarding the accuracy and the value of previous research published in the literature between 2011 and 2018. The second goal is to extensively report on the holistic, experimental outcomes of this domain in relation to current research. It is systematically performed to provide a wealthy image and grounded evidence of the current state of research covering EEG-based control for assistive rehabilitation devices to all the experts and scientists. The third goal is to recognize the gap of knowledge that demands further investigation and to recommend directions for future research in this area.

A muscle-driven approach to restore stepping with an exoskeleton for individuals with paraplegia.

BACKGROUND: Functional neuromuscular stimulation, lower limb orthosis, powered lower limb exoskeleton, and hybrid neuroprosthesis (HNP) technologies can restore stepping in individuals with paraplegia due to spinal cord injury (SCI). However, a self-contained muscle-driven controllable exoskeleton approach based on an implanted neural stimulator to restore walking has not been previously demonstrated, which could potentially result in system use outside the laboratory and viable for long term use or clinical testing. In this work, we designed and evaluated an untethered muscle-driven controllable exoskeleton to restore stepping in three individuals with paralysis from SCI. METHODS: The self-contained HNP combined neural stimulation to activate the paralyzed muscles and generate joint torques for limb movements with a controllable lower limb exoskeleton to stabilize and support the user. An onboard controller processed exoskeleton sensor signals, determined appropriate exoskeletal constraints and stimulation commands for a finite state machine (FSM), and transmitted data over Bluetooth to an off-board computer for real-time monitoring and data recording. The FSM coordinated stimulation and exoskeletal constraints to enable functions, selected with a wireless finger switch user interface, for standing up, standing, stepping, or sitting down. In the stepping function, the FSM used a sensor-based gait event detector to determine transitions between gait phases of double stance, early swing, late swing, and weight acceptance. RESULTS: The HNP restored stepping in three individuals with motor complete paralysis due to SCI. The controller appropriately coordinated stimulation and exoskeletal constraints using the sensor-based FSM for subjects with different stimulation systems. The average range of motion at hip and knee joints during walking were 8.5°-20.8° and 14.0°-43.6°, respectively. Walking speeds varied from 0.03 to 0.06 m/s, and cadences from 10 to 20 steps/min. CONCLUSIONS: A self-contained muscle-driven exoskeleton was a feasible intervention to restore stepping in individuals with paraplegia due to SCI. The untethered hybrid system was capable of adjusting to different individuals' needs to appropriately coordinate exoskeletal constraints with muscle activation using a sensor-driven FSM for stepping. Further improvements for out-of-the-laboratory use should include implantation of plantar flexor muscles to improve walking speed and power assist as needed at the hips and knees to maintain walking as muscles fatigue.