Review of electromyography onset detection methods for real-time control of robotic exoskeletons.

BACKGROUND: Electromyography (EMG) is a classical technique used to record electrical activity associated with muscle contraction and is widely applied in Biomechanics, Biomedical Engineering, Neuroscience and Rehabilitation Robotics. Determining muscle activation onset timing, which can be used to infer movement intention and trigger prostheses and robotic exoskeletons, is still a big challenge. The main goal of this paper was to perform a review of the state-of-the-art of EMG onset detection methods. Moreover, we compared the performance of the most commonly used methods on experimental EMG data. METHODS: A total of 156 papers published until March 2022 were included in the review. The papers were analyzed in terms of application domain, pre-processing method and EMG onset detection method. The three most commonly used methods [Single (ST), Double (DT) and Adaptive Threshold (AT)] were applied offline on experimental intramuscular and surface EMG signals obtained during contractions of ankle and knee joint muscles. RESULTS: Threshold-based methods are still the most commonly used to detect EMG onset. Compared to ST and AT, DT required more processing time and, therefore, increased onset timing detection, when applied on experimental data. The accuracy of these three methods was high (maximum error detection rate of 7.3%), demonstrating their ability to automatically detect the onset of muscle activity. Recently, other studies have tested different methods (especially Machine Learning based) to determine muscle activation onset offline, reporting promising results. CONCLUSIONS: This study organized and classified the existing EMG onset detection methods to create consensus towards a possible standardized method for EMG onset detection, which would also allow more reproducibility across studies. The three most commonly used methods (ST, DT and AT) proved to be accurate, while ST and AT were faster in terms of EMG onset detection time, especially when applied on intramuscular EMG data. These are important features towards movement intention identification, especially in real-time applications. Machine Learning methods have received increased attention as an alternative to detect muscle activation onset. However, although several methods have shown their capability offline, more research is required to address their full potential towards real-time applications, namely to infer movement intention.

Muscle synergies analysis shows altered neural strategies in women with patellofemoral pain during walking.

Several studies suggest that the central nervous system coordinates muscle activation by modulating neural commands directed to groups of muscles combined to form muscle synergies. Individuals with patellofemoral pain (PFP) move differently from asymptomatic individuals. Understanding the neural strategies involved in the execution of tasks such as walking can help comprehend how the movement is planned and better understand this clinical condition. The objective of this study was to compare muscle synergies between women with and without PFP during walking. Eleven women with PFP and thirteen asymptomatic women were assessed using three-dimensional kinematics and electromyography (EMG) while walking at self-selected speed. Kinematics of the trunk, pelvis and lower limbs were analyzed through the Movement Deviation Profile. Muscle synergies were extracted from the EMG signals of eight lower limb muscles collected throughout the whole gait cycle. Kinematic differences between the two groups (p<0.001, z-score = 3.06) were more evident during loading response, terminal stance, and pre-swing. PFP group presented a lower number of muscle synergies (p = 0.037), and greater variability accounted for (VAFtotal) when using 3 (p = 0.017), 4 (p = 0.004), and 5 (p = 0.012) synergies to reconstruct all EMG signals. The PFP group also presented higher VAFmuscle for rectus femoris (p = 0.048) and gastrocnemius medialis (p = 0.019) when considering 4 synergies. Our results suggest that women with PFP show lower motor complexity and deficit in muscle coordination to execute gait, indicating that gait in PFP is the result of different neural commands compared to asymptomatic women.

Mechanical disturbances applied by motorized ankle foot orthosis to adapt ankle muscles activation-A validation study.

Background: Reduced function of ankle muscles usually leads to impaired gait. Motorized ankle foot orthoses (MAFOs) have shown potential to improve neuromuscular control and increase volitional engagement of ankle muscles. In this study, we hypothesize that specific disturbances (adaptive resistance-based perturbations to the planned trajectory) applied by a MAFO can be used to adapt the activity of ankle muscles. The first goal of this exploratory study was to test and validate two different ankle disturbances based on plantarflexion and dorsiflexion resistance while training in standing still position. The second goal was to assess neuromuscular adaptation to these approaches, namely, in terms of individual muscle activation and co-activation of antagonists. Methods: Two ankle disturbances were tested in ten healthy subjects. For each subject, the dominant ankle followed a target trajectory while the contralateral leg was standing still: a) dorsiflexion torque during the first part of the trajectory (Stance Correlate disturbance-StC), and b) plantarflexion torque during the second part of the trajectory (Swing Correlate disturbance-SwC). Electromyography was recorded from the tibialis anterior (TAnt) and gastrocnemius medialis (GMed) during MAFO and treadmill (baseline) trials. Results: GMed (plantarflexor muscle) activation decreased in all subjects during the application of StC, indicating that dorsiflexion torque did not enhance GMed activity. On the other hand, TAnt (dorsiflexor muscle) activation increased when SwC was applied, indicating that plantarflexion torque succeeded in enhancing TAnt activation. For each disturbance paradigm, there was no antagonist muscle co-activation accompanying agonist muscle activity changes. Conclusion: We successfully tested novel ankle disturbance approaches that can be explored as potential resistance strategies in MAFO training. Results from SwC training warrant further investigation to promote specific motor recovery and learning of dorsiflexion in neural-impaired patients. This training can potentially be beneficial during intermediate phases of rehabilitation prior to overground exoskeleton-assisted walking. Decreased activation of GMed during StC might be attributed to the unloaded body weight in the ipsilateral side, which typically decreases activation of anti-gravity muscles. Neural adaptation to StC needs to be studied thoroughly in different postures in futures studies.

Classification of Kinematic and Electromyographic Signals Associated with Pathological Tremor Using Machine and Deep Learning.

Peripheral Electrical Stimulation (PES) of afferent pathways has received increased interest as a solution to reduce pathological tremors with minimal side effects. Closed-loop PES systems might present some advantages in reducing tremors, but further developments are required in order to reliably detect pathological tremors to accurately enable the stimulation only if a tremor is present. This study explores different machine learning (K-Nearest Neighbors, Random Forest and Support Vector Machines) and deep learning (Long Short-Term Memory neural networks) models in order to provide a binary (Tremor; No Tremor) classification of kinematic (angle displacement) and electromyography (EMG) signals recorded from patients diagnosed with essential tremors and healthy subjects. Three types of signal sequences without any feature extraction were used as inputs for the classifiers: kinematics (wrist flexion-extension angle), raw EMG and EMG envelopes from wrist flexor and extensor muscles. All the models showed high classification scores (Tremor vs. No Tremor) for the different input data modalities, ranging from 0.8 to 0.99 for the f1 score. The LSTM models achieved 0.98 f1 scores for the classification of raw EMG signals, showing high potential to detect tremors without any processed features or preliminary information. These models may be explored in real-time closed-loop PES strategies to detect tremors and enable stimulation with minimal signal processing steps.

Floating EMG sensors and stimulators wirelessly powered and operated by volume conduction for networked neuroprosthetics.

BACKGROUND: Implantable neuroprostheses consisting of a central electronic unit wired to electrodes benefit thousands of patients worldwide. However, they present limitations that restrict their use. Those limitations, which are more adverse in motor neuroprostheses, mostly arise from their bulkiness and the need to perform complex surgical implantation procedures. Alternatively, it has been proposed the development of distributed networks of intramuscular wireless microsensors and microstimulators that communicate with external systems for analyzing neuromuscular activity and performing stimulation or controlling external devices. This paradigm requires the development of miniaturized implants that can be wirelessly powered and operated by an external system. To accomplish this, we propose a wireless power transfer (WPT) and communications approach based on volume conduction of innocuous high frequency (HF) current bursts. The currents are applied through external textile electrodes and are collected by the wireless devices through two electrodes for powering and bidirectional digital communications. As these devices do not require bulky components for obtaining power, they may have a flexible threadlike conformation, facilitating deep implantation by injection. METHODS: We report the design and evaluation of advanced prototypes based on the above approach. The system consists of an external unit, floating semi-implantable devices for sensing and stimulation, and a bidirectional communications protocol. The devices are intended for their future use in acute human trials to demonstrate the distributed paradigm. The technology is assayed in vitro using an agar phantom, and in vivo in hindlimbs of anesthetized rabbits. RESULTS: The semi-implantable devices were able to power and bidirectionally communicate with the external unit. Using 13 commands modulated in innocuous 3 MHz HF current bursts, the external unit configured the sensing and stimulation parameters, and controlled their execution. Raw EMG was successfully acquired by the wireless devices at 1 ksps. CONCLUSIONS: The demonstrated approach overcomes key limitations of existing neuroprostheses, paving the way to the development of distributed flexible threadlike sensors and stimulators. To the best of our knowledge, these devices are the first based on WPT by volume conduction that can work as EMG sensors and as electrical stimulators in a network of wireless devices.

Effects of gravity and kinematic constraints on muscle synergies in arm cycling.

Arm cycling is a bimanual motor task used in medical rehabilitation and in sports training. Understanding how muscle coordination changes across different biomechanical constraints in arm cycling is a step toward improved rehabilitation approaches. This exploratory study aims to get new insights on motor control during arm cycling. To achieve our main goal, we used the muscle synergies analysis to test three hypotheses: 1) body position with respect to gravity (sitting and supine) has an effect on muscle synergies; 2) the movement size (crank length) has an effect on the synergistic behavior; 3) the bimanual cranking mode (asynchronous and synchronous) requires different synergistic control. Thirteen able-bodied volunteers performed arm cranking on a custom-made device with unconnected cranks, which allowed testing three different conditions: body position (sitting vs. supine), crank length (10 cm vs. 15 cm), and cranking mode (synchronous vs. asynchronous). For each of the eight possible combinations, subjects cycled for 30 s while electromyography of eight muscles (four from each arm) were recorded: biceps brachii, triceps brachii, anterior deltoid, and posterior deltoid. Muscle synergies in this eight-dimensional muscle space were extracted by nonnegative matrix factorization. Four synergies accounted for over 90% of muscle activation variances in all conditions. Results showed that synergies were affected by body position and cranking mode but practically unaffected by movement size. These results suggest that the central nervous system may employ different motor control strategies in response to external constraints such as cranking mode and body position during arm cycling.NEW & NOTEWORTHY Recent studies analyzed muscle synergies in lower limb cycling. Here, we examine upper limb cycling and specifically the effect of body position with respect to gravity, movement size, and cranking mode on muscle coordination during arm cranking tasks. We show that altered body position and cranking mode affects modular organization of muscle activities. To our knowledge, this is the first study assessing motor control through muscle synergies framework during upper limb cycling with different constraints.