672 results found
Chen C, Yu Y, Sheng X, et al., 2021, Simultaneous and proportional control of wrist and hand movements by decoding motor unit discharges in real time, JOURNAL OF NEURAL ENGINEERING, Vol: 18, ISSN: 1741-2560
Tottrup L, Atashzar SF, Farina D, et al., 2021, Altered evoked low-frequency connectivity from SI to ACC following nerve injury in rats, JOURNAL OF NEURAL ENGINEERING, Vol: 18, ISSN: 1741-2560
Savić AM, Aliakbaryhosseinabadi S, Blicher JU, et al., 2021, Online control of an assistive active glove by slow cortical signals in patients with amyotrophic lateral sclerosis., J Neural Eng, Vol: 18
Objective.A brain-computer interface (BCI) allows users to control external devices using brain signals that can be recorded non-invasively via electroencephalography (EEG). Movement related cortical potentials (MRCPs) are an attractive option for BCI control since they arise naturally during movement execution and imagination, and therefore, do not require an extensive training. This study tested the feasibility of online detection of reaching and grasping using MRCPs for the application in patients suffering from amyotrophic lateral sclerosis (ALS).Approach.A BCI system was developed to trigger closing of a soft assistive glove by detecting a reaching movement. The custom-made software application included data collection, a novel method for collecting the input data for classifier training from the offline recordings based on a sliding window approach, and online control of the glove. Eight healthy subjects and two ALS patients were recruited to test the developed BCI system. They performed assessment blocks without the glove active (NG), in which the movement detection was indicated by a sound feedback, and blocks (G) in which the glove was controlled by the BCI system. The true positive rate (TPR) and the positive predictive value (PPV) were adopted as the outcome measures. Correlation analysis between forehead EEG detecting ocular artifacts and sensorimotor area EEG was conducted to confirm the validity of the results.Main results.The overall median TPR and PPV were >0.75 for online BCI control, in both healthy individuals and patients, with no significant difference across the blocks (NG versus G).Significance.The results demonstrate that cortical activity during reaching can be detected and used to control an external system with a limited amount of training data (30 trials). The developed BCI system can be used to provide grasping assistance to ALS patients.
Jung MK, Muceli S, Rodrigues C, et al., 2021, Intramuscular EMG-driven Musculoskeletal Modelling: Towards Implanted Muscle Interfacing in Spinal Cord Injury Patients., IEEE Trans Biomed Eng, Vol: PP
OBJECTIVE: Surface EMG-driven modelling has been proposed as a means to control assistive devices by estimating joint torques. Implanted EMG sensors have several advantages over wearable sensors but provide a more localized information on muscle activity, which may impact torque estimates. Here, we tested and compared the use of surface and intramuscular EMG measurements for the estimation of required assistive joint torques using EMG driven modelling. METHODS: Four healthy subjects and three incomplete spinal cord injury (SCI) patients performed walking trials at varying speeds. Motion capture marker trajectories, surface and intramuscular EMG, and ground reaction forces were measured concurrently. Subject-specific musculoskeletal models were developed for all subjects, and inverse dynamics analysis was performed for all individual trials. EMG-driven modelling based joint torque estimates were obtained from surface and intramuscular EMG. RESULTS: The correlation between the experimental and predicted joint torques was similar when using intramuscular or surface EMG as input to the EMG-driven modelling estimator in both healthy individuals and patients. CONCLUSION: We have provided the first comparison of non-invasive and implanted EMG sensors as input signals for torque estimates in healthy individuals and SCI patients. SIGNIFICANCE: Implanted EMG sensors have the potential to be used as a reliable input for assistive exoskeleton joint torque actuation.
Ibanez Pereda J, Alessandro DV, John C R, et al., 2021, Only the fastest corticospinal fibers contribute to beta corticomuscular coherence, The Journal of Neuroscience, Vol: 41, Pages: 4867-4879, ISSN: 0270-6474
Human corticospinal transmission is commonly studied using brain stimulation. However, this approach is biased to activity in the fastest conducting axons. It is unclear whether conclusions obtained in this context are representative of volitional activity in mild-to-moderate contractions. An alternative to overcome this limitation may be to study the corticospinal transmission of endogenously generated brain activity. Here we investigate in humans (N=19; of either sex), the transmission speeds of cortical beta rhythms (∼20Hz) traveling to arm (first dorsal interosseous) and leg (tibialis anterior) muscles during tonic mild contractions. For this purpose, we propose two improvements for the estimation of cortico-muscular beta transmission delays. First, we show that the cumulant density (cross-covariance) is more accurate than the commonly-used directed coherence to estimate transmission delays in bidirectional systems transmitting band-limited signals. Second, we show that when spiking motor unit activity is used instead of interference electromyography, cortico-muscular transmission delay estimates are unaffected by the shapes of the motor unit action potentials. Applying these improvements, we show that descending cortico-muscular beta transmission is only 1-2ms slower than expected from the fastest corticospinal pathways. In the last part of our work, we show results from simulations using estimated distributions of the conduction velocities for descending axons projecting to lower motoneurons (from macaque histological measurements) to suggest two scenarios that can explain fast cortico-muscular transmission: either only the fastest corticospinal axons selectively transmit beta activity, or else the entire pool does. The implications of these two scenarios for our understanding of corticomuscular interactions are discussed.
Hug F, Avrillon S, Del Vecchio A, et al., 2021, Analysis of motor unit spike trains estimated from high-density surface electromyography is highly reliable across operators, Journal of Electromyography and Kinesiology, Vol: 58, Pages: 102548-102548, ISSN: 1050-6411
There is a growing interest in decomposing high-density surface electromyography (HDsEMG) into motor unit spike trains to improve knowledge on the neural control of muscle contraction. However, the reliability of decomposition approaches is sometimes questioned, especially because they require manual editing of the outputs. We aimed to assess the inter-operator reliability of the identification of motor unit spike trains. Eight operators with varying experience in HDsEMG decomposition were provided with the same data extracted using the convolutive kernel compensation method. They were asked to manually edit them following established procedures. Data included signals from three lower leg muscles and different submaximal intensities. After manual analysis, 126 ± 5 motor units were retained (range across operators: 119-134). A total of 3380 rate of agreement values were calculated (28 pairwise comparisons × 11 contractions/muscles × 4-28 motor units). The median rate of agreement value was 99.6%. Inter-operator reliability was excellent for both mean discharge rate and time at recruitment (intraclass correlation coefficient > 0.99). These results show that when provided with the same decomposed data and the same basic instructions, operators converge toward almost identical results. Our data have been made available so that they can be used for training new operators.
Farina D, Vujaklija I, Branemark R, et al., 2021, Toward higher-performance bionic limbs for wider clinical use, NATURE BIOMEDICAL ENGINEERING, ISSN: 2157-846X
Natalie M-K, Ibanez Pereda J, Dario F, 2021, Towards a mechanistic approach for the development of non-invasive brain-computer interfaces for motor rehabilitation, The Journal of Physiology, Vol: 599, Pages: 2361-2374, ISSN: 0022-3751
Brain‐computer interfaces (BCIs) designed for motor rehabilitation use brain signals associated with motor‐processing states to guide neuroplastic changes in a state‐dependent manner. These technologies are uniquely positioned to induce targeted and functionally relevant plastic changes in the human motor nervous system. However, while several studies have shown that BCI‐based neuromodulation interventions may improve motor function in patients with lesions in the central nervous system, the neurophysiological structures and processes targeted with the BCI interventions have not been identified. In this review, we first summarize current knowledge of the changes in the central nervous system associated with learning new motor skills. Then, we propose a classification of current BCI paradigms for plasticity induction and motor rehabilitation based on the expected neural plastic changes promoted. This classification proposes four paradigms based on two criteria: the plasticity induction methods and the brain states targeted. The existing evidence regarding the brain circuits and processes targeted with these different BCIs is discussed in detail. The proposed classification aims to serve as a starting point for future studies trying to elucidate the underlying plastic changes following BCI interventions.
Farina D, Mrachacz-Kersting N, 2021, Brain-computer interfaces and plasticity of the human nervous system, JOURNAL OF PHYSIOLOGY-LONDON, Vol: 599, Pages: 2349-2350, ISSN: 0022-3751
Sturma A, Hruby LA, Boesendorfer A, et al., 2021, Prosthetic embodiment and body image changes in patients undergoing bionic reconstruction following brachial plexus injury, Frontiers in Neurorobotics, Vol: 15, Pages: 1-10, ISSN: 1662-5218
Brachial plexus injuries with multiple-root involvement lead to severe and long-lasting impairments in the functionality and appearance of the affected upper extremity. In cases, where biologic reconstruction of hand and arm function is not possible, bionic reconstruction may be considered as a viable clinical option. Bionic reconstruction, through a careful combination of surgical augmentation, amputation, and prosthetic substitution of the functionless hand, has been shown to achieve substantial improvements in function and quality of life. However, it is known that long-term distortions in the body image are present in patients with severe nerve injury as well as in prosthetic users regardless of the level of function. To date, the body image of patients who voluntarily opted for elective amputation and prosthetic reconstruction has not been investigated. Moreover, the degree of embodiment of the prosthesis in these patients is unknown. We have conducted a longitudinal study evaluating changes of body image using the patient-reported Body Image Questionnaire 20 (BIQ-20) and a structured questionnaire about prosthetic embodiment. Six patients have been included. At follow up 2.5–5 years after intervention, a majority of patients reported better BIQ-20 scores including a less negative body evaluation (5 out of 6 patients) and higher vital body dynamics (4 out of 6 patients). Moreover, patients described a strong to moderate prosthesis embodiment. Interestingly, whether patients reported performing bimanual tasks together with the prosthetic hand or not, did not influence their perception of the prosthesis as a body part. In general, this group of patients undergoing prosthetic substitution after brachial plexus injury shows noticeable inter-individual differences. This indicates that the replacement of human anatomy with technology is not a straight-forward process perceived in the same way by everyone opting for it.
Aliakbaryhosseinabadi S, Lontis R, Farina D, et al., 2021, Effect of motor learning with different complexities on EEG spectral distribution and performance improvement, Biomedical Signal Processing and Control, Vol: 66, ISSN: 1746-8094
Motor learning can improve movement performance and behavioral measurements, such as reaction time, by inducing brain plasticity. In this study, we investigated the effect of training with different task complexity on Electroencephalographic (EEG) signals. Two types of training (‘simple’ and ‘complex’) were performed by two groups of healthy volunteers. The complex training group (CTG) performed a trace tracking task using their dominant foot and the simple training group (STG) executed repetitive ankle dorsiflexion in the training phase. Frequency analysis was performed to study the effect of training on EEG signals. In addition, the coherence between paired-channels investigated to represent changes in brain region connectivity. Results revealed that the power in the Beta (15−31 Hz) was significantly reduced while gamma band power (32−80 Hz) was significantly enhanced in the CTG compared to the STG mainly in the frontal, central and centro-parietal channels. Theta power was also increased after training in fronto-central channel. Moreover, performance variations were mainly correlated to the beta and gamma power changes. Finally, the connectivity of gamma and beta band increased significantly particularly between frontal and central region in CTG while connectivity score of theta and delta band decreased after training. These findings confirm that training-induced brain plasticity depends on the complexity of the task, more complexity.
Guo W, Ma C, Wang Z, et al., 2021, Long exposure convolutional memory network for accurate estimation of finger kinematics from surface electromyographic signals., Journal of Neural Engineering, Vol: 18, Pages: 1-12, ISSN: 1741-2552
OBJECTIVE: Estimation of finger kinematics is an important function of an intuitive human-machine interface, such as gesture recognition. Here, we propose a novel deep learning method, named Long Exposure Convolutional Memory Network (LE-ConvMN), and use it to proportionally estimate finger joint angles through surface electromyographic (sEMG) signals. APPROACH: We use a convolution structure to replace the neuron structure of traditional Long Short-Term Memory (LSTM) networks, and use the long exposure data structure which retains the spatial and temporal information of the electrodes as input. The Ninapro database, which contains continuous finger gestures and corresponding sEMG signals was used to verify the efficiency of the proposed deep learning method. The proposed method was compared with LSTM and Sparse Pseudo-input Gaussian Process (SPGP) on this database to predict the 10 main joint angles on the hand based on sEMG. The correlation coefficient (CC) was evaluated using the three methods on eight healthy subjects, and all the methods adopted the root mean square (RMS) features. MAIN RESULTS: The experimental results showed that the average CC, RMSE, NRMSE of the proposed LE-ConvMN method (0.82±0.03,11.54±1.89,0.12±0.013) was significantly higher than SPGP (0.65±0.05, p＜0.001; 15.51±2.82, p＜0.001; 0.16±0.01, p＜0.001) and LSTM (0.64±0.06, p＜0.001; 14.77±3.21, p＜0.001; 0.15±0.02, p=＜0.001). Furthermore, the proposed real-time-estimation method has a computation cost of only approximately 82 ms to output one state of ten joints (average value of 10 tests on TitanV GPU). SIGNIFICANCE: The proposed LE-ConvMN method could efficiently estimate the continuous movement of fingers with sEMG, and its performance is significantly superior to two established deep learning methods.
Enoka RM, Farina D, 2021, Force Steadiness: From Motor Units to Voluntary Actions., Physiology (Bethesda), Vol: 36, Pages: 114-130
Voluntary actions are controlled by the synaptic inputs that are shared by pools of spinal motor neurons. The slow common oscillations in the discharge times of motor units due to these synaptic inputs are strongly correlated with the fluctuations in force during submaximal isometric contractions (force steadiness) and moderately associated with performance scores on some tests of motor function. However, there are key gaps in knowledge that limit the interpretation of differences in force steadiness.
Ting JE, Vecchio AD, Sarma D, et al., 2021, Sensing and decoding the neural drive to paralyzed muscles during attempted movements of a person with tetraplegia using a sleeve array
<jats:title>Abstract</jats:title><jats:p>Motor neurons in the brain and spinal cord convey information about motor intent that can be extracted and interpreted to control assistive devices, such as computers, wheelchairs, and robotic manipulators. However, most methods for measuring the firing activity of single neurons rely on implanted microelectrodes. Although intracortical brain-computer interfaces (BCIs) have been shown to be safe and effective, the requirement for surgery poses a barrier to widespread use. Here, we demonstrate that a wearable sensor array can detect residual motor unit activity in paralyzed muscles after severe cervical spinal cord injury (SCI). Despite generating no observable hand movement, volitional recruitment of motor neurons below the level of injury was observed across attempted movements of individual fingers and overt wrist and elbow movements. Subgroups of motor units were coactive during flexion or extension phases of the task. Single digit movement intentions were classified offline from the EMG power (RMS) or motor unit firing rates with median classification accuracies >75% in both cases. Simulated online control of a virtual hand was performed with a binary classifier to test feasibility of real time extraction and decoding of motor units. The online decomposition algorithm extracted motor units in 1.2 ms, and the firing rates predicted the correct digit motion 88 ± 24% of the time. This study provides the first demonstration of a wearable interface for recording and decoding firing rates of motor neurons below the level of injury in a person with tetraplegia after motor complete SCI.</jats:p><jats:sec><jats:title>Significance Statement</jats:title><jats:p>A wearable electrode array and machine learning methods were used to record and decode myoelectric signals and motor unit firing in paralyzed muscles of a person with motor complete tetraplegia. Motor unit action p
Clarke AK, Atashzar SF, Vecchio AD, et al., 2021, Deep learning for robust decomposition of high-density surface EMG signals, IEEE Transactions on Biomedical Engineering, Vol: 68, Pages: 526-534, ISSN: 0018-9294
Blind source separation (BSS) algorithms, such as gradient convolution kernel compensation (gCKC), can efficiently and accurately decompose high-density surface electromyography (HD-sEMG) signals into constituent motor unit (MU) action potential trains. Once the separation matrix is blindly estimated on a signal interval, it is also possible to apply the same matrix to subsequent signal segments. Nonetheless, the trained separation matrices are sub-optimal in noisy conditions and require that incoming data undergo computationally expensive whitening. One unexplored alternative is to instead use the paired HD-sEMG signal and BSS output to train a model to predict MU activations within a supervised learning framework. A gated recurrent unit (GRU) network was trained to decompose both simulated and experimental unwhitened HD-sEMG signal using the output of the gCKC algorithm. The results on the experimental data were validated by comparison with the decomposition of concurrently recorded intramuscular EMG signals. The GRU network outperformed gCKC at low signal-to-noise ratios, proving superior performance in generalising to new data. Using 12 seconds of experimental data per recording, the GRU performed similarly to gCKC, at rates of agreement of 92.5% (84.5%-97.5%) and 94.9% (88.8%-100.0%) respectively for GRU and gCKC against matched intramuscular sources.
Hug F, Del Vecchio A, Avrillon S, et al., 2021, Muscles from the same muscle group do not necessarily share common drive: evidence from the human triceps surae., Journal of applied physiology (Bethesda, Md. : 1985), Vol: 130, Pages: 342-354, ISSN: 1522-1601
It has been proposed that movements are produced through groups of muscles, or motor modules, activated by common neural commands. However, the neural origin of motor modules is still debated. Here, we used complementary approaches to determine: 1) whether three muscles of the same muscle group [soleus, gastrocnemius medialis (GM), and gastrocnemius lateralis (GL)] are activated by a common neural drive, and 2) whether the neural drive to GM and GL could be differentially modified by altering the mechanical requirements of the task. Eighteen human participants performed an isometric standing heel raise and submaximal isometric plantarflexions (10%, 30%, and 50% of maximal effort). High-density surface electromyography recordings were decomposed into motor unit action potentials and coherence analysis was applied on the motor unit spike trains. We identified strong common drive to each muscle but minimal common drive between the muscles. Further, large between-muscle differences were observed during the isometric plantarflexions, such as a delayed recruitment time of GL compared with GM and soleus motor units and opposite time-dependent changes in the estimates of neural drive to muscles during the torque plateau. Finally, the feet position adopted during the heel-raise task (neutral vs. internally rotated) affected only the GL neural drive with no change for GM. These results provide conclusive evidence that not all anatomically defined synergist muscles are controlled by strong common neural drive. Independent drive to some muscles from the same muscle group may allow for more flexible control to comply with secondary goals such as joint stabilization.NEW & NOTEWORTHY In this study, we demonstrated that the three muscles composing the human triceps surae share minimal common drive during isometric contractions. Our results suggest that reducing the number of effectively controlled degrees of freedom may not always be the strategy used by the central nervous sys
Braecklein M, Ibanez J, Barsakcioglu DY, et al., 2021, Towards human motor augmentation by voluntary decoupling beta activity in the neural drive to muscle and force production, JOURNAL OF NEURAL ENGINEERING, Vol: 18, ISSN: 1741-2560
Avrillon S, Del Vecchio A, Farina D, et al., 2021, Individual differences in the neural strategies to control the lateral and medial head of the quadriceps during a mechanically constrained task., Journal of applied physiology (Bethesda, Md. : 1985), Vol: 130, Pages: 269-281, ISSN: 1522-1601
The interindividual variability in the neural drive sent from the spinal cord to muscles is largely unknown, even during highly constrained motor tasks. Here, we investigated individual differences in the strength of neural drive received by the vastus lateralis (VL) and vastus medialis (VM) during an isometric task. We also assessed the proportion of common neural drive within and between these muscles. Twenty-two participants performed a series of submaximal isometric knee extensions at 25% of their peak torque. High-density surface electromyography recordings were decomposed into motor unit action potentials. Coherence analyses were applied on the motor unit spike trains to assess the degree of neural drive that was shared between motor neurons. Six participants were retested ∼20 mo after the first session. The distribution of the strength of neural drive between VL and VM varied between participants and was correlated with the distribution of normalized interference electromyography (EMG) signals (r > 0.56). The level of within- and between-muscle coherence varied across individuals, with a significant positive correlation between these two outcomes (VL: r = 0.48; VM: r = 0.58). We also observed a large interindividual variability in the proportion of muscle-specific drive, that is, the drive unique to each muscle (VL range: 6%-83%, VM range: 6%-86%). All the outcome measures were robust across sessions, providing evidence that the individual differences did not depend solely on the variability of the measures. Together, these results demonstrate that the neural strategies to control the VL and VM muscles widely vary across individuals, even during a constrained task.NEW & NOTEWORTHY We observed that the distribution of the strength of neural drive between the vastus lateralis and vastus medialis during a single-joint isometric task varied across participants. Also, we observed that the proportion of ne
Jiang X, Liu X, Fan J, et al., 2021, Enhancing IoT Security via Cancelable HD-sEMG-based Biometric Authentication Password, Encoded by Gesture, IEEE Internet of Things Journal
Enhancing information security via reliable user authentication in wireless body area network (WBAN)-based Internet of Things (IoT) applications has attracted increasing attention. The noncancelability of traditional biometrics (e.g. fingerprint) for user authentication increases the privacy disclosure risks once the biometric template is exposed, because users cannot volitionally create a new template. In this work, we propose a cancelable biometric modality based on high-density surface electromyogram (HD-sEMG) encoded by hand gesture password, for user authentication. HD-sEMG signals (256 channels) were acquired from the forearm muscles when users performed a prescribed gesture password, forming their biometric token. Thirty four alternative hand gestures in common daily use were studied. Moreover, to reduce the data acquisition and transmission burden in IoT devices, an automatically generated password-specific channel mask was employed to reduce the number of active channels. HD-sEMG biometrics were also robust with reduced sampling rate, further reducing power consumption. HD-sEMG biometrics achieved a low equal error rate (EER) of 0.0013 when impostors entered a wrong gesture password, as validated on 20 subjects. Even if impostors entered the correct gesture password, the HD-sEMG biometrics still achieved an EER of 0.0273. If the HD-sEMG biometric template was exposed, users could cancel it by simply changing it to a new gesture password, with an EER of 0.0013. To the best of our knowledge, this is the first study to employ HD-sEMG signals under common daily hand gestures as biometric tokens, with training and testing data acquired on different days.
Mouchoux J, Carisi S, Dosen S, et al., 2021, Artificial Perception and Semiautonomous Control in Myoelectric Hand Prostheses Increases Performance and Decreases Effort, IEEE Transactions on Robotics, ISSN: 1552-3098
Dexterous control of upper limb prostheses with multiarticulated wrists/hands is still a challenge due to the limitations of myoelectric man–machine interfaces. Multiple factors limit the overall performance and usability of these interfaces, such as the need to control degrees of freedom sequentially and not concurrently, and the inaccuracies in decoding the user intent from weak or fatigued muscles. In this article, we developed a novel man–machine interface that endows a myoelectric prosthesis (MYO) with artificial perception, estimation of user intention, and intelligent control (MYO–PACE) to continuously support the user with automation while preparing the prosthesis for grasping. We compared the MYO–PACE against state-of-the-art myoelectric control (pattern recognition) in laboratory and clinical tests. For this purpose, eight able-bodied and two amputee individuals performed a standard clinical test consisting of a series of manipulation tasks (portion of the SHAP test), as well as a more complex sequence of transfer tasks in a cluttered scene. In all tests, the subjects not only completed the trials faster using the MYO–PACE but also achieved more efficient myoelectric control. These results demonstrate that the implementation of advanced perception, context interpretation, and autonomous decision-making into active prostheses improves control dexterity. Moreover, it also effectively supports the user by speeding up the preshaping phase of the movement and decreasing muscle use.
Urh F, Strnad D, Clarke A, et al., 2021, On the Need for Spatial Whitening of High-Density Surface Electromyograms in Motor Unit Identification by Neural Networks, Pages: 915-922, ISSN: 1680-0737
In recent years, many new decomposition methods for identification of motor unit (MU) firings from high-density surface electromyograms (HDEMG) have been developed, with recent attempts focused on the use of different neural networks (NN). In this study, we evaluated the need for HDEMG signal whitening in NN-based MU identification. For this purpose, we analyzed the learning efficiency of two different types of NN, namely dense NN and long short-term memory (LSTM) NN, on the same HDEMG signals, with and without spatial whitening applied to them. All the HDEMG signals used were simulated with advanced HDEMG simulator, providing a full control of MU firing patterns and MU characteristics in our test environment. Spatial whitening of HDEMG signals significantly improved the precision of MU identification, regardless of the type of NN tested. For dense NN, precision of identified MU increased from 32.2 ± 20.2% to 93.1 ± 8.7%, whereas miss rate decreased from 48.4 ± 23.9% to 12.0 ± 13.3% when whitening of HDEMG signals was employed. For LSTM NN the precision of MU identification increased from 59.7 ± 19.7% to 99.4 ± 2.0% whereas miss rate decreased from 43.1 ± 22.3% to 12.7 ± 9.7% with whitening.
Rahimian E, Zabihi S, Asif A, et al., 2021, FS-HGR: Few-Shot Learning for Hand Gesture Recognition via Electromyography., IEEE Trans Neural Syst Rehabil Eng, Vol: 29, Pages: 1004-1015
This work is motivated by the recent advances in Deep Neural Networks (DNNs) and their widespread applications in human-machine interfaces. DNNs have been recently used for detecting the intended hand gesture through the processing of surface electromyogram (sEMG) signals. Objective: Although DNNs have shown superior accuracy compared to conventional methods when large amounts of data are available for training, their performance substantially decreases when data are limited. Collecting large datasets for training may be feasible in research laboratories, but it is not a practical approach for real-life applications. The main objective of this work is to design a modern DNN-based gesture detection model that relies on minimal training data while providing high accuracy. Methods: We propose the novel Few-Shot learning- Hand Gesture Recognition (FS-HGR) architecture. Few-shot learning is a variant of domain adaptation with the goal of inferring the required output based on just one or a few training observations. The proposed FS-HGR generalizes after seeing very few observations from each class by combining temporal convolutions with attention mechanisms. This allows the meta-learner to aggregate contextual information from experience and to pinpoint specific pieces of information within its available set of inputs. Data Source & Summary of Results: The performance of FS-HGR was tested on the second and fifth Ninapro databases, referred to as the DB2 and DB5, respectively. The DB2 consists of 50 gestures (rest included) from 40 healthy subjects. The Ninapro DB5 contains data from 10 healthy participants performing a total of 53 different gestures (rest included). The proposed approach for the Ninapro DB2 led to 85.94% classification accuracy on new repetitions with few-shot observation (5-way 5-shot), 81.29% accuracy on new subjects with few-shot observation (5-way 5-shot), and 73.36% accuracy on new gestures with few-shot observation (5-way 5-shot). Moreover, the
Tottrup L, Atashzar SF, Farina D, et al., 2020, Nerve injury decreases hyperacute resting-state connectivity between the anterior cingulate and primary somatosensory cortex in anesthetized rats, IEEE Transactions on Neural Systems and Rehabilitation Engineering, Vol: 28, Pages: 2691-2698, ISSN: 1534-4320
A better understanding of neural pain processing and of the development of pain over time, is critical to identify objective measures of pain and to evaluate the effect of pain alleviation therapies. One issue is, that the brain areas known to be related to pain processing are not exclusively responding to painful stimuli, and the neuronal activity is also influenced by other brain areas. Functional connectivity reflects synchrony or covariation of activation between groups of neurons. Previous studies found changes in connectivity days or weeks after pain induction. However, less in known on the temporal development of pain. Our objective was therefore to investigate the interaction between the anterior cingulate cortex (ACC) and primary somatosensory cortex (SI) in the hyperacute (minute) and sustained (hours) response in an animal model of neuropathic pain. Intra-cortical local field potentials (LFP) were recorded in 18 rats. In 10 rats the spared nerve injury model was used as an intervention. The intra-cortical activity was recorded before, immediately after, and three hours after the intervention. The interaction was quantified as the calculated correlation and coherence. The results from the intervention group showed a decrease in correlation between ACC and SI activity, which was most pronounced in the hyperacute phase but a longer time frame may be required for plastic changes to occur. This indicated that both SI and ACC are involved in hyperacute pain processing.
Del Vecchio A, Sylos-Labini F, Mondi V, et al., 2020, Spinal motoneurons of the human newborn are highly synchronized during leg movements, SCIENCE ADVANCES, Vol: 6, ISSN: 2375-2548
Gardner M, Mancero Castillo C, Wilson S, et al., 2020, A multimodal intention detection sensor suite for shared autonomy of upper-limb robotic prostheses, Sensors, Vol: 20, ISSN: 1424-8220
Neurorobotic augmentation (e.g., robotic assist) is now in regular use to support individuals suffering from impaired motor functions. A major unresolved challenge, however, is the excessive cognitive load necessary for the human–machine interface (HMI). Grasp control remains one of the most challenging HMI tasks, demanding simultaneous, agile, and precise control of multiple degrees-of-freedom (DoFs) while following a specific timing pattern in the joint and human–robot task spaces. Most commercially available systems use either an indirect mode-switching configuration or a limited sequential control strategy, limiting activation to one DoF at a time. To address this challenge, we introduce a shared autonomy framework centred around a low-cost multi-modal sensor suite fusing: (a) mechanomyography (MMG) to estimate the intended muscle activation, (b) camera-based visual information for integrated autonomous object recognition, and (c) inertial measurement to enhance intention prediction based on the grasping trajectory. The complete system predicts user intent for grasp based on measured dynamical features during natural motions. A total of 84 motion features were extracted from the sensor suite, and tests were conducted on 10 able-bodied and 1 amputee participants for grasping common household objects with a robotic hand. Real-time grasp classification accuracy using visual and motion features obtained 100%, 82.5%, and 88.9% across all participants for detecting and executing grasping actions for a bottle, lid, and box, respectively. The proposed multimodal sensor suite is a novel approach for predicting different grasp strategies and automating task performance using a commercial upper-limb prosthetic device. The system also shows potential to improve the usability of modern neurorobotic systems due to the intuitive control design.
Sturma A, Stamm T, Hruby LA, et al., 2020, Rehabilitation of high upper limb amputees after Targeted Muscle Reinnervation, Journal of Hand Therapy, ISSN: 0894-1130
STUDY DESIGN: This is a Delphi study based on a scoping literature review. INTRODUCTION: Targeted muscle reinnervation (TMR) enables patients with high upper limb amputations to intuitively control a prosthetic arm with up to six independent control signals. Although there is a broad agreement regarding the importance of structured motor learning and prosthetic training after such nerve transfers, to date, no evidence-based protocol for rehabilitation after TMR exists. PURPOSE OF THE STUDY: We aimed at developing a structured rehabilitation protocol after TMR surgery after major upper limb amputation. The purpose of the protocol is to guide clinicians through the full rehabilitation process, from presurgical patient education to functional prosthetic training. METHODS: European clinicians and researchers working in upper limb prosthetic rehabilitation were invited to contribute to a web-based Delphi study. Within the first round, clinical experts were presented a summary of recent literature and were asked to describe the rehabilitation steps based on their own experience and scientific evidence. The second round was used to refine these steps, while the importance of each step was rated within the third round. RESULTS: Experts agreed on a rehabilitation protocol that consists of 16 steps and starts before surgery. It is based on two overarching principles, namely the necessity of multiprofessional teamwork and a careful selection and education of patients within the rehabilitation team. Among the different steps in therapy, experts rated the training with electromyographic biofeedback as the most important one. DISCUSSION: Within this study, a first rehabilitation protocol for TMR patients based on a broad experts' consensus and relevant literature could be developed. The detailed steps for rehabilitation start well before surgery and prosthetic fitting, and include relatively novel interventions as motor imagery and biofeedback. Future studies need to further inve
Martinez-Valdes E, Negro F, Falla D, et al., 2020, Inability to increase the neural drive to muscle is associated with task failure during submaximal contractions., J Neurophysiol, Vol: 124, Pages: 1110-1121
We investigated changes in motor unit (MU) behavior and vasti-muscle contractile properties during sustained submaximal fatiguing contractions with a new time-domain tracking technique to understand the mechanisms responsible for task failure. Sixteen participants performed a nonfatiguing 15-s isometric knee extension at 50% of the maximum voluntary (MVC) torque, followed by a 30% MVC sustained contraction until exhaustion. Two grids of 64 surface electromyography electrodes were placed over vastus medialis and lateralis. Signals were decomposed into MU discharge times and the MUs from the 30% MVC sustained contraction were followed until task failure by overlapping decomposition intervals. These MUs were then tracked between 50% and 30% MVC. During the sustained fatiguing contraction, MUs of the two muscles decreased their discharge rate until ∼40% of the endurance time, referred to as the reversal time, and then increased their discharge rate until task failure. This reversal in firing behavior predicted total endurance time and was matched by opposite changes in twitch force (increase followed by a decrease). Despite the later increase in MU firing rates, peak discharge rates at task failure did not reach the frequency attained during a nonfatiguing 50% MVC contraction. These results show that changes in MU firing properties are influenced by adjustments in contractile properties during the course of the contraction, allowing the identification of two phases. Nevertheless, the contraction cannot be sustained, possibly because of progressive motoneuron inhibition/decreased excitability, as the later increase in firing rate saturates at a much lower frequency compared with a higher-force nonfatiguing contraction.NEW & NOTEWORTHY Motor unit firing and contractile properties during a submaximal contraction until failure were assessed with a new tracking technique. Two distinct phases in firing behavior were observed, which compensated for changes in twitch ar
Atashzar SF, Huang H-Y, Duca FD, et al., 2020, Energetic Passivity Decoding of Human Hip Joint for Physical Human-Robot Interaction, IEEE ROBOTICS AND AUTOMATION LETTERS, Vol: 5, Pages: 5953-5960, ISSN: 2377-3766
Nuccio S, Del Vecchio A, Casolo A, et al., 2020, Muscle fiber conduction velocity in the vastus lateralis and medialis muscles of soccer players after ACL reconstruction, Scandinavian Journal of Medicine and Science in Sports, Vol: 30, Pages: 1976-1984, ISSN: 0905-7188
The neural factors underlying the persistency of quadriceps weakness after anterior cruciate ligament reconstruction (ACLR) have been only partially explained. This study examined muscle fiber conduction velocity (MFCV) as an indirect parameter of motor unit recruitment strategies in the vastus lateralis (VL) and medialis (VM) muscles of soccer players with ACLR. High‐density surface electromyography (HDsEMG) was acquired from VL and VM in nine soccer players (22.7 ± 2.9 years; BMI: 22.08 ± 1.72 kg·m−2; 7.7 ± 2.2 months post‐surgery). Voluntary muscle force and the relative myoelectrical activity from the reconstructed and contralateral sides were recorded during linearly increasing isometric knee extension contractions up to 70% of maximal voluntary isometric force (MVIF). The relation of MFCV and force was examined by linear regression analysis at the individual subject level. The initial (intercept), peak (MFCV70), and rate of change (slope) of MFCV related to force were compared between limbs and muscles. The MVIF was lower in the reconstructed side than in the contralateral side (−%20.5; P < .05). MFCV intercept was similar among limbs and muscles (P > .05). MFCV70 and MFCV slope were lower in the reconstructed side compared to the contralateral for both VL (−28.5% and −10.1%, respectively; P < .001) and VM (−22.6% and −8.1%, respectively; P < .001). The slope of MFCV was lower in the VL than VM, but only in the reconstructed side (−12.4%; P < .001). These results suggest possible impairments in recruitment strategies of high‐threshold motor units (HTMUs) as well as deficits in sarcolemmal excitability, fiber diameter, and discharge rate of knee extensor muscles following ACLR.
Pascual Valdunciel A, Gonzalez-Sanchez M, Muceli S, et al., 2020, Intramuscular stimulation of muscle afferents attains prolonged tremor reduction in essential tremor patients., IEEE Transactions on Biomedical Engineering, Vol: PP, ISSN: 0018-9294
This study proposes and clinically tests intramuscular electrical stimulation below motor threshold to achieve prolonged reduction of wrist flexion/extension tremor in Essential Tremor (ET) patients. The developed system consisted of an intramuscular thin-film electrode structure that included both stimulation and electromyography (EMG) recording electrodes, and a control algorithm for the timing of intramuscular stimulation based on EMG (closed-loop stimulation). Data were recorded from nine ET patients with wrist flexion/extension tremor recruited from the Gregorio Maran Hospital (Madrid, Spain). Patients participated in two experimental sessions comprising: 1) sensory stimulation of wrist flexors/extensors via thin-film multichannel intramuscular electrodes; and 2) surface stimulation of the nerves innervating the same target muscles. For each session, four of these patients underwent random 60-s trials of two stimulation strategies for each target muscle: 1) selective and adaptive timely stimulation (SATS) - based on EMG of the antagonist muscle; and 2) continuous stimulation (CON) of target muscles. Two patients underwent SATS stimulation trials alone while the other three underwent CON stimulation trials alone in each session. Kinematics of wrist, elbow, and shoulder, together with clinical scales, were used to assess tremor before, right after, and 24 h after each session. Intramuscular SATS achieved, on average, 32% acute (during stimulation) tremor reduction on each trial, while continuous stimulation augmented tremorgenic activity. Furthermore, tremor reduction was significantly higher using intramuscular than surface stimulation. Prolonged reduction of tremor amplitude (24 h after the experiment) was observed in four patients. These results showed acute and prolonged (24 h) tremor reduction using a minimally invasive neurostimulation technology based on SATS of primary sensory afferents of wrist muscles. This strategy might open the possibility of an alte
This data is extracted from the Web of Science and reproduced under a licence from Thomson Reuters. You may not copy or re-distribute this data in whole or in part without the written consent of the Science business of Thomson Reuters.