Allied rehabilitation using caregiver-mediated exercises combined with telerehabilitation for stroke (ARMed4Stroke): A randomised controlled trial.

OBJECTIVE: To assess the added value of caregiver-mediated exercises combined with telerehabilitation in addition to usual care compared to usual care alone on the self-reported mobility outcome after subacute stroke. DESIGN: Multicentre, observer-blinded, parallel randomised controlled trial. An off-site researcher allocated treatments using minimisation. SETTING: Four rehabilitation centres in the Netherlands. PARTICIPANTS: Forty-one patient-caregiver dyads within 3 months poststroke. INTERVENTION: Eight-week blended care program with caregiver-mediated mobility exercises for 2.5 h per week supported by telerehabilitation and four face-to-face sessions in addition to usual care. MAIN MEASURES: Self-reported mobility domain of the Stroke Impact Scale postintervention. Secondary outcomes were functional outcome, dyads' psychosocial wellbeing, care transition to the community postintervention and after 6 months. RESULTS: Forty-one dyads (21 intervention, 20 control) were randomised, and 37 (N = 18; N = 19) were analysed following intention-to-treat. The Stroke Impact Scale mobility was not significantly different between groups postintervention (B 0.8, 95% CI -6.8-8.5, p = 0.826). The secondary outcomes, namely, (a) caregivers' quality of life postintervention (p = 0.013), (b) caregivers' symptoms of depression postintervention (p = 0.025), and (c) independence in leisurely activities at 6 months (p = 0.024), showed significant benefits in favour of caregiver-mediated exercises with telerehabilitation. A significant difference favouring controls was found in self-reported muscle strength at 6 months (p = 0.002). CONCLUSIONS: Caregiver-mediated exercises combined with telerehabilitation yielded no differential effect on our primary outcome self-reported mobility. Although the trial is underpowered, current findings are in line with previous trials. Future studies should further explore beneficial effects of caregiver involvement in stroke rehabilitation targeting psychosocial wellbeing.

Testing and evaluation of lower limb prosthesis prototypes in people with a transfemoral amputation: a scoping review on research protocols.

BACKGROUND: When developing new lower limb prostheses, prototypes are tested to obtain insights into the performance. However, large variations between research protocols may complicate establishing the potential added value of newly developed prototypes over other prostheses. OBJECTIVE: This review aims at identifying participant characteristics, research protocols, reference values, aims, and corresponding outcome measures used during prosthesis prototype testing on people with a transfemoral amputation. METHODS: A systematic search was done on PubMed and Scopus from 2000 to December 2020. Articles were included if testing was done on adults with transfemoral or knee disarticulation amputation; testing involved walking with a non-commercially available prototype leg prosthesis consisting of at least a knee component; and included evaluations of the participants' functioning with the prosthesis prototype. RESULTS: From the initial search of 2027 articles, 48 articles were included in this review. 20 studies were single-subject studies and 4 studies included a cohort of 10 or more persons with a transfemoral amputation. Only 5 articles reported all the pre-defined participant characteristics that were deemed relevant. The familiarization time with the prosthesis prototype prior to testing ranged from 5 to 10 min to 3 months; in 25% of the articles did not mention the extent of the familiarization period. Mobility was most often mentioned as the development or testing aim. A total of 270 outcome measures were identified, kinetic/kinematic gait parameters were most often reported. The majority of outcome measures corresponded to the mobility aim. For 48% of the stated development aims and 4% of the testing aims, no corresponding outcome measure could be assigned. Results indicated large inconsistencies in research protocols and outcome measures used to validate pre-determined aims. CONCLUSIONS: The large variation in prosthesis prototype testing and reporting calls for the development of a core set of reported participant characteristics, testing protocols, and specific and well-founded outcome measures, tailored to the various aims and development phases. The use of such a core set can give greater insights into progress of developments and determine which developments have additional benefits over the state-of-the-art. This review may contribute as initial input towards the development of such a core set.

Adaptive Lower Limb Pattern Recognition for Multi-Day Control.

Pattern recognition in EMG-based control systems suffer from increase in error rate over time, which could lead to unwanted behavior. This so-called concept drift in myoelectric control systems could be caused by fatigue, sensor replacement and varying skin conditions. To circumvent concept drift, adaptation strategies could be used to retrain a pattern recognition system, which could lead to comparable error rates over multiple days. In this study, we investigated the error rate development over one week and compared three adaptation strategies to reduce the error rate increase. The three adaptation strategies were based on entropy, on backward prediction and a combination of backward prediction and entropy. Ten able-bodied subjects were measured on four measurement days while performing gait-related activities. During the measurement electromyography and kinematics were recorded. The three adaptation strategies were implemented and compared against the baseline error rate and against adaptation using the ground truth labels. It can be concluded that without adaptation the baseline error rate increases significantly from day 1 to 2, but plateaus on day 2, 3 and 7. Of the three tested adaptation strategies, entropy based adaptation showed the smallest increase in error rate over time. It can be concluded that entropy based adaptation is simple to implement and can be considered a feasible adaptation strategy for lower limb pattern recognition.

Influence of a user-adaptive prosthetic knee on quality of life, balance confidence, and measures of mobility: a randomised cross-over trial.

OBJECTIVE: To study the influence of a transition from a non-microprocessor controlled to the Rheo Knee(®) II on quality of life, balance confidence and measures of mobility. DESIGN: Randomised crossover trial. SETTING: Research department of a rehabilitation centre. SUBJECTS: Persons with a transfemoral amputation or knee disarticulation (n=10). INTERVENTIONS: Participants were assessed with their own non-microprocessor controlled knee and with the Rheo Knee(®) II. The low-profile Vari-Flex with EVO foot was installed in both knee conditions, followed by eight weeks of acclimatisation. The order in which knees were tested was randomised. MAIN MEASURES: Prosthesis Evaluation Questionnaire with addendum, Activities-specific Balance Confidence scale, Timed "up & go" test, Timed up and down stairs test, Hill Assessment Index, Stairs Assessment Index, Standardized Walking Obstacle Course and One Leg Balance test. RESULTS: Significant higher scores were found for the Rheo Knee(®) II on the Residual Limb Health subscale of the Prosthesis Evaluation Questionnaire when compared to the non-microprocessor controlled prosthetic knee (median [interquartile range] resp. 86.67 [62.21-93.08] and 68.71 [46.15-94.83]; P=0.047) In addition, participants needed significantly more steps to complete an obstacle course when walking with the Rheo Knee(®) II compared to the non-microprocessor controlled prosthetic knee (median [interquartile range] resp. 23.50 [19.92-26.25] and 22.17 [19.50-25.75]; P=0.041). On other outcome measures, no significant differences were found. CONCLUSIONS: Transition towards the Rheo Knee(®) II had little effect on the studied outcome measures.

Effect of chemodenervation of the rectus femoris muscle in adults with a stiff knee gait due to spastic paresis: a systematic review with a meta-analysis in patients with stroke.

OBJECTIVE: To determine the effect of motor branch block (MBB) or neuromuscular block (NMB) of the rectus femoris on knee kinematics during swing, functional outcome, and energy cost in adults with spastic paresis presenting a stiff knee gait. DATA SOURCES: PubMed, Embase, CINAHL, and Cochrane Library were searched. Studies were collected up to February 26, 2013. Reference lists were additionally scrutinized. STUDY SELECTION: No restrictions were applied regarding study design. Patients were adults suffering from a central neurological disorder. Interventions had to include MBB or NMB. Outcome measures had to include knee kinematics during the swing phase. Study selection was independently performed by 2 reviewers. DATA EXTRACTION: Two reviewers independently assessed the methodological quality of included studies. Data on kinematics, functional outcome, and energy cost from patients with stroke were extracted from the total population and when possible pooled. DATA SYNTHESIS: A total of 9 articles describing 12 different studies were included. Knee kinematics (peak knee flexion or knee range) during swing improved significantly in all the included studies. The average increase in peak knee flexion varied from 1.9° to 15.4°. Data pooling of peak knee flexion in patients with stroke showed a significant improvement of 7.37° (P=.000) in NMB studies and of 9.35° (P=.002) in MBB studies. Data pooling of knee velocity at toe-off showed a significant improvement of 53.01°/s in NMB studies. In MBB studies, this improvement was not significant. Data pooling of knee range of motion, functional outcomes, and energy cost showed no significant difference. CONCLUSIONS: According to this review, chemodenervation of the rectus femoris shows a significant improvement in peak knee flexion during swing. The effect on functional outcomes and energy cost is still unclear.

A robot-based hybrid lower limb system for Assist-As-Needed rehabilitation of stroke patients: Technical evaluation and clinical feasibility.

BACKGROUND: Although early rehabilitation is important following a stroke, severely affected patients have limited options for intensive rehabilitation as they are often bedridden. To create a system for early rehabilitation of lower extremities in these patients, we combined the robotic manipulator ROBERT® with electromyography (EMG)-triggered functional electrical stimulation (FES) and developed a novel user-driven Assist-As-Needed (AAN) control. The method is based on a state machine able to detect user movement capability, assessed by the presence of an EMG-trigger and the movement velocity, and provide different levels of assistance as required by the patient (no support, FES only, and simultaneous FES and mechanical assistance). METHODS: To technically validate the system, we tested 10 able-bodied participants who were instructed to perform specific behaviors to test the system states while conducting knee extension and ankle dorsal flexion exercises. The system was also tested on two stroke patients to establish its clinical feasibility. RESULTS: The technical validation showed that the state machine correctly detected the participants' behavior and activated the target AAN state in more than 96% of the exercise repetitions. The clinical feasibility test showed that the system successfully recognized the patients' movement capacity and activated assistive states according to their needs providing the minimal level of support required to exercise successfully. CONCLUSIONS: The system was technically validated and preliminarily proved clinically feasible. The present study shows that the novel system can be used to deliver exercises with a high number of repetitions while engaging the participants' residual capabilities through the AAN strategy.

Combining robotics and functional electrical stimulation for assist-as-needed support of leg movements in stroke patients: A feasibility study.

PURPOSE: Rehabilitation technology can be used to provide intensive training in the early phases after stroke. The current study aims to assess the feasibility of combining robotics and functional electrical stimulation (FES), with an assist-as-needed approach to support actively-initiated leg movements in (sub-)acute stroke patients. METHOD: Nine subacute stroke patients performed repetitions of ankle dorsiflexion and/or knee extension movements, with and without assistance. The assist-as-needed algorithm determined the amount and type of support needed per repetition. The number of repetitions and range of motion with and without assistance were compared with descriptive statistics. Fatigue scores were obtained using the visual analogue scale (score 0-10). RESULTS: Support was required in 44 % of the repetitions for ankle dorsiflexion and in 5 % of the repetitions of knee extension, The median fatigue score was 2.0 (IQR: 0.2) and 4.0 (IQR: 1.5) for knee and ankle, respectively, indicating mild to moderate perceived fatigue. CONCLUSION: This study demonstrated the feasibility of assist-as-needed assistance through combined robotic and FES support of leg movements in stroke patients. It proved particularly useful for ankle dorsiflexion. Future research should focus on implementing this approach in a clinical setting, to assess clinical applicability and potential effects on leg function.

Comparison of muscle activity patterns of transfemoral amputees and control subjects during walking.

BACKGROUND: Only few studies have looked at electromyography (EMG) during prosthetic gait. Differences in EMG between normal and prosthetic gait for stance and swing phase were never separately analyzed. These differences can give valuable information if and how muscle activity changes in prosthetic gait. METHODS: In this study EMG activity during gait of the upper leg muscles of six transfemoral amputees, measured inside their own socket, was compared to that of five controls. On and off timings for stance and swing phase were determined together with the level of co-activity and inter-subject variability. RESULTS AND CONCLUSIONS: Gait phase changes in amputees mainly consisted of an increased double support phase preceding the prosthetic stance phase. For the subsequent (pre) swing phase the main differences were found in muscle activity patterns of the prosthetic limb, more muscles were active during this phase and/or with prolonged duration. The overall inter-subject variability was larger in amputees compared to controls.

The Effect of Botulinum Toxin Type A Injection in the Rectus Femoris in Stroke Patients Walking With a Stiff Knee Gait: A Randomized Controlled Trial.

BACKGROUND: Over activity of the rectus femoris is often cited as a main cause for stiff knee gait (SKG). Botulinum toxin (BoNT) can be used to reduce this over activity. Inconsistent results for the effect of BoNT injections were found in literature which can possibly be explained by the study design as these were uncontrolled or non-randomized studies. OBJECTIVE: To conduct a randomized controlled trial (RCT) to investigate the effect of botulinum toxin type A (BoNT-A) injections in the rectus femoris on gait kinematics and functional outcome in adult stroke patients. METHODS: Twenty-six participants were included in this triple-blind cross-over RCT. The intervention consisted of an injection with BoNT-A. Placebo is an injection with saline. Besides knee and hip kinematics, functional outcomes were measured. RESULTS: Comparison of the effect of BoNT-A injection to placebo injection showed a significant increase in peak knee flexion and knee range of motion of 6.7° and 4.8° respectively. There was no difference in hip kinematics. In functional outcomes, only the 6 Minute Walking Test showed a significant increase of 18.3 m. CONCLUSIONS: BoNT-A injections in the rectus femoris is a valuable treatment option for stroke patients walking with a SKG to improve knee kinematics. To study the effect on functional outcome more research is necessary with different functional outcome measures that can capture the effect in kinematics. It is important to use kinematic measurements to demonstrate effects in quality of movement that are not captured by commonly used functional outcome measurements post stroke.Clinical Trial Registration: https://trialsearch.who.int/Trial2.aspx?TrialID=NTR2169.

Database of lower limb kinematics and electromyography during gait-related activities in able-bodied subjects.

This data descriptor describes the Roessingh Research & Development-MyLeg database for activity prediction (MyPredict), containing three data sets. These data sets contain data from 55 able-bodied subjects, mean age 24 ± 2 years, measured in 85 measurement sessions. Measurement sessions consisted of trials containing sitting, standing, overground walking, stair ascent, stair descent, ramp ascent, ramp descent, walking on uneven terrain and walking in simulated confined spaces. Subjects were measured using eight inertial measurement units in combination with different types of sEMG. Recorded kinematics consisted of joint angles, sensor accelerations, angular velocity, orientation and virtual marker positions. sEMG was recorded using bipolar sEMG, multi-array sEMG or a combination of both. All data showed excellent correlation with other online available data sets. The data reported in this descriptor forms a solid basis for research into myoelectric pattern recognition, myoelectric control development and electromyography to be used in data-driven applications.

Head support in wheelchairs (scoping review): state-of-the-art and beyond.

BACKGROUND: Many wheelchair users experience disabilities in stabilising and positioning of the head. For these users, adequate head support is required. Although several types of head supports are available, further development of these systems is needed to improve functionality and quality of life, especially for the group of severely challenged users. For this group, user needs have not been clearly established. In this article, we provide an overview of the state-of-the-art in wheelchair mounted head supports and associated scientific evidence in order to identify requirements for the next generation of head support systems. MATERIALS AND METHODS: A scoping review was performed including scientific literature (PubMed/Scopus), patents (Espacenet/Google Scholar) and commercial information. Types of head support and important system characteristics for future head support systems were proposed from consultations with wheelchair users (n = 3), occupational therapists (n = 3) and an expert panel. RESULTS: Forty scientific papers, 90 patents and 80 descriptions of commercial devices were included in the scoping review. The identified head support systems were categorised per head support type. Only limited scientific clinical evidence with respect to the effectiveness of existing head support systems was found. From the user and expert consultations, a need was identified for personalised head support systems that intuitively combine changes in sitting and head position with continuous optimal support of the head to accommodate severely challenged users. CONCLUSIONS: This study presents the state-of-the-art in head support systems. Additionally, several important system characteristics are introduced that provide guidance for the development and improvement of head supports.Implications for rehabilitationEspecially for the group of severely challenged wheelchair users, current head support systems require further development to improve their users' quality of life.The desired system characteristics which are discussed in this review are an important step in the definition of requirements for the next generation of head supports.

Amputation and prosthetics of the lower extremity: The 2020 Dutch evidence-based multidisciplinary guideline.

BACKGROUND: Lower-limb amputations are rare but debilitating events in the lives of affected persons. Treatment of persons with amputation inherently involves many different health care professions at different stages leading to and after an amputation. There are prevailing clinical questions within the work field related to different facets of care including peri/postoperative aspects, prosthetic components, rehabilitation treatment, and health care processes. OBJECTIVES: To provide an up-to-date multidisciplinary evidence-based guideline for health care professionals involved in the treatment of persons with lower-limb amputation in the Netherlands. METHODS: Identification of key questions in a focus group, systematic review of the evidence (up to March 2019, using Embase and MEDLINE databases), and weighing considerations, culminating in clinical recommendations. RESULTS: Twelve key questions were formulated. Recommendations of two key questions were upheld in line with the previous 2012 guideline. Ten systematic literature searches were performed, leading to the inclusion of 59 studies. CONCLUSION: A summary of evidence-based conclusions, considerations, and recommendations of the 2020 guideline is presented.

Multi-Day EMG-Based Knee Joint Torque Estimation Using Hybrid Neuromusculoskeletal Modelling and Convolutional Neural Networks.

Proportional control using surface electromyography (EMG) enables more intuitive control of a transfemoral prosthesis. However, EMG is a noisy signal which can vary over time, giving rise to the question what approach for knee torque estimation is most suitable for multi-day control. In this study we compared three different modelling frameworks to estimate knee torque in non-weight-bearing situations. The first model contained a convolutional neural network (CNN) which mapped EMG to knee torque directly. The second used a neuromusculoskeletal model (NMS) which used EMG, muscle tendon unit lengths and moment arms to compute knee torque. The third model (Hybrid) used a CNN to map EMG to specific muscle activation, which was used together with NMS components to compute knee torque. Multi-day measurements were conducted on ten able-bodied participants who performed non-weight bearing activities. CNN had the best performance in general and on each day (Normalized Root Mean Squared Error (NRMSE) 9.2 ± 4.4%). The Hybrid model (NRMSE 12.4 ± 3.4%) was able to outperform NMS (NRMSE 14.3 ± 4.2%). The NMS model showed no significant difference between measurement days. The CNN model and Hybrid models had significant performance differences between the first day and all other days. CNNs are suited for multi-day torque estimation in terms of error rate, outperforming the other two model types. NMS was the only model type which was robust over all days. This study investigated the behavior of three model types over multiple days, giving insight in the most suited modelling approach for multi-day torque estimation to be used in prosthetic control.

Synchronization of wearable motion capture and EMG measurement systems.

Synchronization of motion capture systems with other modalities in out-of-the-lab settings is not trivial. Various synchronization methods exist, such as using servers or transistor-transistor-logic pulses. However, not all measurement set-ups allow for such synchronization methods. Therefore, we have developed and validated an acceleration based post-measurement method to synchronize an IMU based motion capture system and an EMG measurement device. On top of the thigh IMU an additional accelerometer was placed which was connected to the analog input of the EMG device. By applying cross-correlation continuously, the similarities in the measured acceleration by the two measurement systems can be used for synchronization. We performed a validation measurement on seven able-bodied subjects and tested various correlation window sizes in hour long measurements in an out of the lab setting. It can be concluded that the developed method works for different activities when a suitable window length is chosen for cross-correlation. If no other options are available for synchronization, this correlation based method using an additional accelerometer is a viable option.

Detection thresholds for electrostimulation combined with robotic leg support in sub-acute stroke patients.

Stroke is one of the leading causes of disability in adults in the European Union. It often leads to motor impairments, such as a hemiparetic lower extremity. Research indicates that early task-specific and intensive training promotes neuroplasticity and leads to recovery and/or compensation. One way to provide intensive training early after a stroke is via robot-supported training. A rehabilitation robot was designed by Life Science Robotics (Aalborg, Denmark) that can provide continuous repetitive movements of the hip, knee, and/or ankle in e.g., a lying position. In order to emphasize active contribution by the patient, actively triggered electrical stimulation (via muscle activation) can be combined with robotic assistance. The current study aims to compare different threshold estimation methods for detection of movement intention from muscle activity for actively triggered electrical stimulation during robot-supported leg movement in stroke patients. Three sub-acute stroke patients were included for a single measurement session. They performed knee extension and/or ankle dorsal flexion with four different threshold estimation methods to assess the intention detection threshold to initiate electrostimulation. The thresholds were based on the resting level of muscle activity (of m. rectus femoris or m. tibialis anterior) plus two or three times the standard deviation of the average resting value, or the resting level plus 5% or 10% of the peak muscle activity during a maximal voluntary contraction. The results showed that the method based on the resting muscle activity plus two times the standard deviation was the most stable across the three included stroke patients. This method had a detection success rate of 86.7% and was experienced as moderately comfortable. In conclusion, performing knee extension and/or ankle dorsal flexion with electromyography triggered electrostimulation is feasible in sub-acute stroke patients. Muscle activity-triggered electrostimulation combined with robotic support based on a threshold of resting levels plus two times the standard deviation seems to detect movement initiation most consistently in this small sample of sub-acute stroke patients.

Does the Duncan-Ely test predict abnormal activity of the rectus femoris in stroke survivors with a stiff knee gait?

OBJECTIVE: To determine the diagnostic value of the Duncan-Ely test in predicting abnormal rectus femoris activity during gait in stroke survivors walking with a stiff knee gait. DESIGN: Cross-sectional diagnostic study. SUBJECTS: A total of 95 patients with chronic stroke. METHODS: During physical examination, the Duncan-Ely test was performed and scored. Surface electromyography of the rectus femoris was then recorded during dynamic gait. To determine the diagnostic value, the results of the Duncan-Ely test and surface electromyography recordings (gold standard) were compared. RESULTS: The Duncan-Ely test had a sensitivity of 73%, a specificity of 29%, a positive predictive value of 60%, and a negative predictive value of 42%. The area under the curve was 0.488 ([AQ1] CI 0.355-0.621, p = 0.862), showing that the Duncan-Ely test is not better than random guessing. CONCLUSION: The Duncan-Ely test has no predictive value for determining abnormal activity of the rectus femoris during gait. Using this test can lead to incorrect identification of abnormal rectus femoris activity, which might hamper the selection of optimal treatment options. We recommend stopping use of the Duncan-Ely test to predict rectus femoris overactivity during swing, and instead use surface electromyography.

Adaptation strategies of the lower extremities of patients with a transtibial or transfemoral amputation during level walking: a systematic review.

OBJECTIVE: To describe adaptation strategies in terms of joint power or work in the amputated and intact leg of patients with a transtibial (TT) or transfemoral (TF) amputation. DATA SOURCES: MEDLINE, CINAHL, Physiotherapy Evidence Database, Embase, and the Cochrane Register of Controlled Trials were searched. Studies were collected up to November 1, 2010. Reference lists were additionally scrutinized. STUDY SELECTION: Studies were included when they presented joint power or work and compared (1) the amputated and intact legs, (2) the amputated leg and a referent leg, or (3) the intact leg and a referent leg. Eligibility was independently assessed by 2 reviewers. A total of 13 articles were identified. DATA EXTRACTION: Data extraction was performed using standardized forms of the Cochrane Collaboration. Methodologic quality was independently assessed using the Downs and Black instrument by 2 reviewers. The possibility of data pooling was examined. Significant differences found in studies that could not be pooled are also presented. DATA SYNTHESIS: Significant results (P<.05). For work TT, for the concentric work total stance phase knee, the amputated was less than the intact/referent side, and the referent was less than the intact side. For the eccentric knee extensor (K1) phase, the amputated was less than the intact side, and the intact was greater than the referent side. For the concentric knee extensor (K2) phase, the amputated/referent was less than the intact side. For the concentric work total stance phase hip, the amputated/intact was greater than the referent side. For the concentric hip extensor (H1) phase, the amputated/intact was greater than the referent side. For power TT, for the peak power generation stance phase knee, the amputated was less than the referent side. For peak power generation swing phase knee, the amputated was less than the referent side. For the eccentric knee flexor (K4) phase, the amputated was less than the intact side. For the eccentric hip flexor (H2) phase, the amputated was greater than the intact side. For work TF, for the concentric plantar flexor (A2) phase, the referent was less than the intact side. For the H1 phase, the referent was less than the intact side. For the H2 phase, the amputated was greater than the intact/referent side, and the referent was greater than the intact side. For power TF, for the K2 phase, the referent was less than the intact side. Sensitivity analysis did not alter the conclusions. CONCLUSIONS: Adaptations were seen in the amputated and intact legs. TT and TF use remarkably similar adaptation strategies at the level of the hip to compensate for the loss of plantar flexion power and facilitate forward progression. At the knee level, adaptations differed between TT and TF.

Genetic Algorithm for Feature Selection in Lower Limb Pattern Recognition.

Choosing the right features is important to optimize lower limb pattern recognition, such as in prosthetic control. EMG signals are noisy in nature, which makes it more challenging to extract useful information. Many features are used in the literature, which raises the question which features are most suited for use in lower limb myoelectric control. Therefore, it is important to find combinations of best performing features. One way to achieve this is by using a genetic algorithm, a meta-heuristic capable of searching vast feature spaces. The goal of this research is to demonstrate the capabilities of a genetic algorithm and come up with a feature set that has a better performance than the state-of-the-art feature set. In this study, we collected a dataset containing ten able-bodied subjects who performed various gait-related activities while measuring EMG and kinematics. The genetic algorithm selected features based on the performance on the training partition of this dataset. The selected feature sets were evaluated on the remaining test set and on the online benchmark dataset ENABL3S, against a state-of-the-art feature set. The results show that a feature set based on the selected features of a genetic algorithm outperforms the state-of-the-art set. The overall error decreased up to 0.54% and the transitional error by 2.44%, which represent a relative decrease in overall errors up to 11.6% and transitional errors up to 14.1%, although these results were not significant. This study showed that a genetic algorithm is capable of searching a large feature space and that systematic feature selection shows promising results for lower limb myoelectric control.

A Comparison of Control Strategies in Commercial and Research Knee Prostheses.

GOAL: To provide an overview of control strategies in commercial and research microprocessor-controlled prosthetic knees (MPKs). METHODS: Five commercially available MPKs described in patents, and five research MPKs reported in scientific literature were compared. Their working principles, intent recognition, and walking controller were analyzed. Speed and slope adaptability of the walking controller was considered as well. RESULTS: Whereas commercial MPKs are mostly passive, i.e., do not inject energy in the system, and employ heuristic rule-based intent classifiers, research MPKs are all powered and often utilize machine learning algorithms for intention detection. Both commercial and research MPKs rely on finite state machine impedance controllers for walking. Yet while commercial MPKs require a prosthetist to adjust impedance settings, scientific research is focused on reducing the tunable parameter space and developing unified controllers, independent of subject anthropometrics, walking speed, and ground slope. CONCLUSION: The main challenges in the field of powered, active MPKs (A-MPKs) to boost commercial viability are first to demonstrate the benefit of A-MPKs compared to passive MPKs or mechanical non-microprocessor knees using biomechanical, performance-based and patient-reported metrics. Second, to evaluate control strategies and intent recognition in an uncontrolled environment, preferably outside the laboratory setting. And third, even though research MPKs favor sophisticated algorithms, to maintain the possibility of practical and comprehensible tuning of control parameters, considering optimal control cannot be known a priori. SIGNIFICANCE: This review identifies main challenges in the development of A-MPKs, which have thus far hindered their broad availability on the market.

Occurrence and Type of Adverse Events During the Use of Stationary Gait Robots-A Systematic Literature Review.

Robot-assisted gait training (RAGT) devices are used in rehabilitation to improve patients' walking function. While there are some reports on the adverse events (AEs) and associated risks in overground exoskeletons, the risks of stationary gait trainers cannot be accurately assessed. We therefore aimed to collect information on AEs occurring during the use of stationary gait robots and identify associated risks, as well as gaps and needs, for safe use of these devices. We searched both bibliographic and full-text literature databases for peer-reviewed articles describing the outcomes of stationary RAGT and specifically mentioning AEs. We then compiled information on the occurrence and types of AEs and on the quality of AE reporting. Based on this, we analyzed the risks of RAGT in stationary gait robots. We included 50 studies involving 985 subjects and found reports of AEs in 18 of those studies. Many of the AE reports were incomplete or did not include sufficient detail on different aspects, such as severity or patient characteristics, which hinders the precise counts of AE-related information. Over 169 device-related AEs experienced by between 79 and 124 patients were reported. Soft tissue-related AEs occurred most frequently and were mostly reported in end-effector-type devices. Musculoskeletal AEs had the second highest prevalence and occurred mainly in exoskeleton-type devices. We further identified physiological AEs including blood pressure changes that occurred in both exoskeleton-type and end-effector-type devices. Training in stationary gait robots can cause injuries or discomfort to the skin, underlying tissue, and musculoskeletal system, as well as unwanted blood pressure changes. The underlying risks for the most prevalent injury types include excessive pressure and shear at the interface between robot and human (cuffs/harness), as well as increased moments and forces applied to the musculoskeletal system likely caused by misalignments (between joint axes of robot and human). There is a need for more structured and complete recording and dissemination of AEs related to robotic gait training to increase knowledge on risks. With this information, appropriate mitigation strategies can and should be developed and implemented in RAGT devices to increase their safety.