Search

The increasing need for sharing workspace and interactive physical tasks between robots and humans has raised concerns regarding safety of such operations. In this regard, controllable clutches have shown great potential for addressing important safety concerns at the hardware level by separating the high-impedance actuator from the end effector by providing the power transfer from electromagnetic source to the human. However, the existing clutches suffer from high power consumption and large-weight, which make them undesirable from the design point of view. In this paper, for the first time, the design and development of a novel, lightweight, and low-power torque-adjustable rotary clutch using electroadhesive materials is presented. The performance of three different pairs of clutch plates is investigated in the context of the smoothness and quality of output torque. The performance degradation issue due to the polarization of the insulator is addressed through the utilization of an alternating current waveform activation signal. Moreover, the effect of the activation frequency on the output torque and power consumption of the clutch is investigated. Finally, a time-dependent model for the output torque of the clutch is presented, and the performance of the clutch was evaluated through experiments, including physical human-robot interaction. The proposed clutch offers a torque to power consumption ratio that is six times better than commercial magnetic particle clutches. The proposed clutch presents great potential for developing safe, lightweight, and low-power physical human-robot interaction systems, such as exoskeletons and robotic walkers., submitted to IEEE T-MECH, 11 pages, 14 figures

Objective evaluation of physiological responses using non-invasive methods for the assessment of vocal performance and voice disorders has attracted great interest. This paper, for the first time, aims to implement and evaluate perilaryngeal-cranial functional muscle networks. The study investigates the variations in topographical characteristics of the network and the corresponding ability to differentiate vocal tasks.Twelve surface electromyography (sEMG) signals were collected bilaterally from six perilaryngeal and cranial muscles. Data were collected from eight subjects (four females) without a known history of voice disorders. The proposed muscle network is composed of pairwise coherence between sEMG recordings. The network metrics include (a) network degree and (b) weighted clustering coefficient (WCC).The varied phonation tasks showed the median degree, and WCC of the muscle network ascend monotonically, with a high effect size ( |rPerilaryngeal-cranial functional muscle network was proposed in this paper. The study showed that the functional muscle network could robustly differentiate the vocal tasks while the classic assessment of muscle activation fails to differentiate.For the first time, we demonstrate the power of a perilaryngeal-cranial muscle network as a neurophysiological window to vocal performance. In addition, the study also discovers tasks with the highest network involvement, which may be utilized in the future to monitor voice disorders and rehabilitation.

In this article, we thoroughly analyze the effect of the single-tendon and antagonistic tendons actuation on the tension loss of multisegment tendon-driven continuum manipulators (TD-CMs) with irregular geometry. To this end, we propose a generic analytical modeling approach and an iterative algorithm that can solve the unknown correlation between the distributed friction force, tendons’ tension transmission loss, and planar deformation behavior of TD-CMs during tendons’ pulling and releasing phases. The proposed generic model solely relies on known input tendons’ tensions and does not require a priori knowledge of the manipulator’s shape and/or other experimental conditions. To investigate the influence of the actuation type on the tension loss and deformation behavior of TD-CMs, we fabricated two different TD-CMs and performed various simulation and experimental studies with the single-tendon and antagonistic tensions actuation. The obtained results indicate the importance of considering the effect of the distributed friction force and actuation type on the tension(s) loss of multisegment TD-CMs. Moreover, it clearly demonstrates the efficacy and accuracy of the proposed model in providing insights and understanding of tension transmission process in various types of actuations in multisegment TD-CMs with irregular geometry.

Patients with both phonotraumatic and non-phonotraumatic dysphonia commonly present with vocal hyperfunction, defined as excessive perilaryngeal muscle activity and characterized by muscular pain and strain in the neck, increased vocal effort, and vocal fatigue. The inability to reliably measure vocal hyperfunction is a barrier to adequate evaluation and treatment of hyperfunctional voice disorders. We have recently demonstrated that the perilaryngeal functional muscle network can be a novel sensitive neurophysiological window to vocal performance in vocally healthy subjects. In this paper, for the first time, we evaluate the performance and symmetry of functional perilaryngeal muscle networks in three patients with voice disorders. Surface electromyography signals were recorded from twelve sensors (six on each side of the neck) using the wireless Trigno sEMG system (Delsys Inc., Natick, MA). Patient 1 was diagnosed with primary muscle tension dysphonia, Patient 2 was diagnosed with unilateral vocal fold paresis, and Patient 3 was diagnosed with age-related glottal insufficiency. This paper reports altered functional connectivity and asymmetric muscle network scan behavior in all three patients when compared with a cohort of eight healthy subjects. Our approach quantifies synergistic network activity to interrogate coordination of perilaryngeal and surrounding muscles during voicing and potentialdiscoordinationof the muscle network for dysphonic conditions. Asymmetry in muscle networks is proposed here as a biomarker for monitoring vocal hyperfunction.

ObjectiveFunctional muscle network analysis has attracted a great deal of interest in recent years, promising high sensitivity to changes of intermuscular synchronicity, studied mostly for healthy subjects and recently for patients living with neurological conditions (e.g., those caused by stroke). Despite the promising results, the between- and within-session reliability of the functional muscle network measures are yet to be established. Here, for the first time, we question and evaluate the test-retest reliability of non-parametric lower-limb functional muscle networks for controlled and lightly-controlled tasks, i.e., sit-to-stand, and over-the-ground walking, respectively, in healthy subjects.MethodFifteen subjects (eight females) were included over two sessions on two different days. The muscle activity was recorded using 14 surface electromyography (sEMG) sensors. The intraclass correlation coefficient (ICC) of the within-session and between-session trials was quantified for the various network metrics, including degree and weighted clustering coefficient. In order to compare with common classical sEMG measures, the reliabilities of the root mean square (RMS) of sEMG and the median frequency (MDF) of sEMG were also calculated.ResultsThe ICC analysis revealed superior between-session reliability for muscle networks, with statistically significant differences when compared to classic measures.Conclusion and SignificanceThis paper proposed that the topographical metrics generated from functional muscle network can be reliably used for multi-session observations securing high reliability for quantifying the distribution of synergistic intermuscular synchronicities of both controlled and lightly controlled lower limb tasks. In addition, the low number of sessions required by the topographical network metrics to reach reliable measurements indicates the potential as biomarkers during rehabilitation.

This paper, for the first time, compares the behaviors of nonlinear versus linear muscle networks in decoding hidden peripheral synergistic neural patterns during dynamic functional tasks. In this paper, we report a case study during which one healthy subject conducts a series of four lower limb repetitive tasks. Specifically, the paper focuses on tasks that involve the right knee joint, including walking, sit-tostand, stepping, and drop-jump. Twelve muscles were recorded using the Delsys Trigno system. The linear muscle network was generated using coherence analysis, and the nonlinear network was generated using Spearman’s correlation. The results show that the degree, clustering coefficient, and global efficiency of the muscle network have the highest value among tasks in the linear domain for the walking task, while a low linear synergistic network behavior for the sit-to-stand is observed. On the other hand, the results show that the nonlinear functional muscle network decodes high connectivity (degree) and clustering coefficient and efficiency for the sit-tostand when compared with other tasks. We have also developed a two-dimensional functional connectivity plane composed of linear and nonlinear features and shown that it can span the lower-limb dynamic task space. The results of this paper for the first time highlight the importance of observing both linear and nonlinear connectivity patterns, especially for complex dynamic tasks. It should also be noted that through a simultaneous EEG recording (using BrainVision System), we have shown that, indeed, cortical activity may indirectly explain highly-connected nonlinear muscle network for the sit-to-stand task, highlighting the importance of nonlinear muscle network as a neurophysiological window of observation beyond the periphery.

The movement-related cortical potential (MRCP) is a low-frequency component of the electroencephalography (EEG) signal recorded from the motor cortex and its neighboring cortical areas. Since the MRCP encodes motor intention and execution, it may be utilized as an interface between patients and neurorehabilitation technologies. This study investigates the EEG signal recorded from the Cz electrode to discriminate between four levels of rate of force development (RFD) of the tibialis anterior muscle. For classification, three feature sets were evaluated to describe the EEG traces. These were (i)MRCP morphological characteristicsin theδ-band such as amplitude and timing, (ii)MRCP statistical characteristicsin theδ-band such as mean, standard deviation, and kurtosis, and (iii)wideband time-frequency featuresin the 0.5-90 Hz range. Using a support vector machine for classification, the four levels of RFD were classified with a mean (SD) accuracy of 82% (7%) accuracy when using the time-frequency feature space, and with an accuracy of 75% (12%) when using the MRCP statistical characteristics. It was also observed that some of the key features from the statistical and morphological sets responded monotonically to the intensity of the RFD. Examples are slope and standard deviation in the (0, 1)s window for the statistical, andmin1andminnfor the morphological sets. This monotonical response of features explains the observed performance of theδ-band MRCP and corresponding high discriminative power. Results from temporal analysis considering the pre-movement phase ((-3, 0)s) and three windows of the post-movement phase ((0, 1)s, (1, 2)s, and (2, 3)s)) suggest that the complete MRCP waveform represents high information content regarding the planning, execution, duration, and ending of the isometric dorsiflexion task using the tibialis anterior muscle. Results shed light on the role ofδ-band in translating to motor command, with potential applications in neural engineering systems.

The spectrotemporal information content of surface electromyography has shown strong potential in predicting the intended motor command. During the last decade, with accelerated exploitation of powerful deep-learning techniques aligned with advancements in active prostheses and neurorobots, a great deal of interest has been drawn to the development of intelligent myoelectric prostheses with an ultimate resolution of upper-limb gestures prediction. Recent research involves Deep CNNs, RNNs, and hybrid frameworks, which have shown promising results. However, deep-learning models have almost always been challenged by the structural complexity, the large number of trainable parameters, concerns of overfitting, and prolonged training time, which complicate the practicality and limit the outcomes. In this letter, for the first time, we propose temporal-dilation in the LSTM module of a hybrid Deepnet model for sEMG-based gesture detection, hypothesizing improved accuracy and training agility. We also analyze the effect of dilation-aggressiveness. We conduct systematic and statistical analysis on the efficacy of the proposed approach in comparison to recent literature, including our previous work. this letter shows that the proposed temporally-dilated LSTM model wins over the recent deep-learning techniques in terms of accuracy, and more significantly, it reduces the training time while increasing the convergence speed, with the ultimate goal of maximizing practicality and translational value for neurorobotic systems.

Soft robots have shown great potential for manufacturing exoskeletons, prostheses, and surgical robots. In this paper, we propose the concept of programmable soft robotics and will experimentally evaluate the performance in the context of continuum mechanisms. The proposed novel concept is motivated by the mechanical shape of RNA molecules which has a single-stranded polymeric molecule with a sugar-phosphate backbone and nitrogenous bases. The shape of the RNA and the type, location, and characteristics of the bases define the coded information. Due to the complexity of RNA, the proposed robot cannot be considered a “bio-inspired” design. Instead, we indirectly utilize the concept of encoding sequences and introduce a new family of soft continuum robots based on a novel design of 3D printable “mechanical library” and “embedded functions” to be implemented on the backbone structure for mechanical programming. Through structural coding of the bases, the paper proposed a wide range of continuum robots. The system has the potential to be scaled up for multiple degrees of freedom (DOF), while the dexterity and range can be structurally programmed. A set of three soft continuum systems are designed, simulated, manufactured. The performance is evaluated by comparing simulations and experiments. We observed that actuators have different hysteresis ranging from 7.50% to 38.36% (on average) with a standard deviation ranging from 5.56% to 40.72%. The results highlight the effect of inherent pneumatic delay causing the hysteresis loops, which should be considered for control.

Pathological hand tremor (PHT) is a common symptom of Parkinson’s disease (PD) and essential tremor (ET), which affects manual targeting, motor coordination, and movement kinetics. Effective treatment and management of the symptoms relies on the correct and in-time diagnosis of the affected individuals, where the characteristics of PHT serve as an imperative metric for this purpose. Due to the overlapping features of the corresponding symptoms, however, a high level of expertise and specialized diagnostic methodologies are required to correctly distinguish PD from ET. In this work, we propose the data-driven $$\text {NeurDNet}$$ NeurDNet model, which processes the kinematics of the hand in the affected individuals and classifies the patients into PD or ET. $$\text {NeurDNet}$$ NeurDNet is trained over 90 hours of hand motion signals consisting of 250 tremor assessments from 81 patients, recorded at the London Movement Disorders Centre, ON, Canada. The $$\text {NeurDNet}$$ NeurDNet outperforms its state-of-the-art counterparts achieving exceptional differential diagnosis accuracy of $$95.55\%$$ 95.55 % . In addition, using the explainability and interpretability measures for machine learning models, clinically viable and statistically significant insights on how the data-driven model discriminates between the two groups of patients are achieved.

In this letter, we propose a novel adaptive iterative stabilization method for physical human-(tele)robot interaction, named Discrete Windowed-Energy Variable Structure Passivity Signature Control (DWE-VSPSC). The proposed stabilizer is capable of adaptively translating the knowledge domain regarding the capacity of the user's biomechanics in absorbing physical interaction energy to reduce the transparency distortion (induced for stabilization) while enhancing system performance through a flexible design of stabilizer. The contributions of this letter are: (1) the design of a novel discrete adaptive stabilizer with proof of stability; allowing for digital implementation of the intelligent algorithm; (2) introducing the concept of “windowed energy stabilization” for the proposed variable structure passivity-signature controller in order to allow for tuning the energy behavior of the interconnected system while making a balance between conservatism and agility of the system; (3) relaxing any assumption on the passivity behavior of the environment while being able to handle stochastic variable network delays without any restriction on the rate of change of delay or the delay. The mathematical design of the stabilizer is provided alongside the stability proof. Also, the performance of the system is evaluated using a systematically-designed grid simulation study for a large range of delay and environmental impedances ranging from passive to non-passive behaviors. A direct application of the outcome is for telerobotic rehabilitation and telerobotic surgery.

This letter investigates the feasibility of a generalizable solution for human-robot interfaces through peripheral multichannel Electromyography (EMG) recording. We propose a tangential approach in comparison to the literature to minimize the need for (re)calibration of the system for new users. The proposed algorithm decodes the signal space and detects the common underlying global neurophysiological components, which can be detected robustly across various users, minimizing the need for retraining and (re)calibration. The research question is how to go beyond techniques that detect a high number of gestures for a given individual (which requires extensive calibration) and achieve an algorithm that can detect a lower number of classes but without the need for (re)calibration. The outcomes of this letter address a challenge affecting the usability and acceptance of advanced myoelectric prostheses. For this, the paper proposes an explainable generalizable hybrid deep learning architecture that incorporates CNN and LSTM. We also utilize the GradCAM analysis to explain and optimize the structure of the generalized model, securing higher computational performance whiles proposing a shallower design.

In this paper, we introduce a new mode of mechanomyography (MMG) signal capture for enhancing the performance of human-machine interfaces (HMIs) through modulation of normal pressure at the sensor location. Utilizing this novel approach, increased MMG signal resolution is enabled by a tunable degree of freedom normal to the sensor-skin contact area. We detail the mechatronic design, experimental validation, and user study of an armband with embedded acoustic sensors demonstrating this capacity. The design is motivated by the nonlinear viscoelasticity of the tissue, which increases with the normal surface pressure. This, in theory, results in higher conductivity of mechanical waves and hypothetically allows to interface with deeper muscle; thus, enhancing the discriminative information context of the signal space. Ten subjects (seven able-bodied and three trans-radial amputees) participated in a study consisting of the classification of hand gestures through MMG while increasing levels of contact force were administered. Four MMG channels were positioned around the forearm and placed over the flexor carpi radialis, brachioradialis, extensor digitorum communis, and flexor carpi ulnaris muscles. A total of 852 spectrotemporal features were extracted (213 features per each channel) and passed through a Neighborhood Component Analysis (NCA) technique to select the most informative neurophysiological subspace of the features for classification. A linear support vector machine (SVM) then classified the intended motion of the user. The results indicate that increasing the normal force level between the MMG sensor and the skin can improve the discriminative power of the classifier, and the corresponding pattern can be user-specific. These results have significant implications enabling embedding MMG sensors in sockets for prosthetic limb control and HMI.

Going beyond the traditional sparse multichannel peripheral human-machine interface that has been used widely in neurorobotics, high-density surface electromyography (HD-sEMG) has shown significant potential for decoding upper-limb motor control. We have recently proposed heterogeneous temporal dilation of LSTM in a deep neural network architecture for a large number of gestures (>60), securing spatial resolution and fast convergence. However, several fundamental questions remain unanswered. One problem targeted explicitly in this paper is the issue of “electrode shift,” which can happen specifically for high-density systems and during doffing and donning the sensor grid. Another real-world problem is the question of transient versus plateau classification, which connects to the temporal resolution of neural interfaces and seamless control. In this paper, for the first time, we implement gesture prediction on the transient phase of HD-sEMG data while robustifying the human-machine interface decoder to electrode shift. For this, we propose the concept of deep data augmentation for transient HD-sEMG. We show that without using the proposed augmentation, a slight shift of 10mm may drop the decoder’s performance to as low as 20%. Combining the proposed data augmentation with a 3D Convolutional Neural Network (CNN), we recovered the performance to 84.6% while securing a high spatiotemporal resolution, robustifying to the electrode shift, and getting closer to large-scale adoption by the end-users, enhancing resiliency.

Sensory information is critical for motor coordination. However, understanding sensorimotor integration is complicated, especially in individuals with nervous system impairment. This research presents a novel functional biomarker, based on a nonlinear network graph of muscle connectivity, called InfoMuNet, to quantify the role of sensory information in motor performance. Thirty-two individuals with post-stroke hemiparesis performed a grasp-and-lift task while muscle activities were measured using eight surface electromyography (sEMG) sensors. Subjects performed the task with their affected hand before and after exposure to the sensory stimulation elicited by performing the task with the less-affected hand. For the first time, this work shows that InfoMuNet robustly quantifies functional muscle connectivity improvements in the affected hand after exposure of the less-affected side to sensory information. >90% of the subjects conformed with the improvement resulting from this sensory exposure. InfoMuNet also shows high sensitivity to tactile, kinesthetic, and visual input alterations at the subject level, highlighting the potential use in precision rehabilitation interventions.

Electromyography (EMG) is a source of neural information for controlling neuroprosthetic devices. To enhance the information content of conventional bipolar EMG, high-density EMG systems include tens to hundreds of closely spaced electrodes that non-invasively record the electrical activity of muscles with high spatial resolution. Despite the advantages of relying on multiple signal sources, however, variations in electrode-skin contact impedance and noise remain challenging for multichannel myocontrol systems. These spatial and temporal non-stationarities negatively impact the control accuracy and therefore substantially limit the clinical viability of high-density EMG techniques. Here, we propose an adaptive algorithm for automatic artefact/noise detection and attenuation for high-density EMG control. The method infers the presence of noise in each EMG channel by spectro-temporal measures of signal similarity. These measures are then used for establishing a scoring system based on an adaptive weighting and reinforcement formulation. The method was experimentally tested as a pre-processing step for a multi-class discrimination problem of 4-digit activation. The approach was proven to enhance the discriminative information content of high-density EMG signals, as well as to attenuate non-stationary artefacts, with improvements in accuracy and robustness of the classification.

A new approach is presented for localizing the Subthalamic Nucleus (STN) during Deep Brain Stimulation (DBS) surgery based on microelectrode recordings (MERs). DBS is an accepted treatment for individuals living with Parkinson’s Disease (PD). This surgery involves implantation of a permanent electrode inside the STN to deliver electrical current. Since the STN is a very small region inside the brain, accurate placement of an electrode is a challenging task for the surgical team. Prior to placement of the permanent electrode, microelectrode recordings of brain activity are used intraoperatively to localize the STN. The placement of the electrode and the success of the therapy depend on this location. In this paper, an objective approach is implemented to help the surgical team in localizing the STN. This is achieved by processing the MER signals and extracting features during the surgery to be used in a Machine Learning (ML) algorithm for defining the neurophysiological borders of the STN. For this purpose, a new classification approach is proposed with the goal of detecting both the dorsal and the ventral borders of the STN during the surgical procedure. Results collected from 100 PD patients in this study, show that by calculating and extracting wavelet transformation features from MER signals and using a data-driven computational deep neural network model, it is possible to detect the borders of the STN with an accuracy of 92%. The proposed method can be implemented in real-time during the surgery to model the neurophysiological nonlinearity along the path of the electrode trajectory during insertion., We present new techniques to assist neurosurgeons during deep brain stimulation (DBS) surgery by providing accurate, real-time localization of the subthalamic nucleus (STN). A combination of discrete wavelet transformation features and a deep neural network algorithm are suggested as a highly accurate approach to localize the STN during DBS surgery with an accuracy of 92%.

While tremendous advances in visual and auditory realism have been made for virtual and augmented reality (VR/AR), introducing a plausible sense of physicality into the virtual world remains challenging. Closing the gap between real-world physicality and immersive virtual experience requires a closed interaction loop: applying user-exerted physical forces to the virtual environment and generating haptic sensations back to the users. However, existing VR/AR solutions either completely ignore the force inputs from the users or rely on obtrusive sensing devices that compromise user experience. By identifying users' muscle activation patterns while engaging in VR/AR, we design a learning-based neural interface for natural and intuitive force inputs. Specifically, we show that lightweight electromyography sensors, resting non-invasively on users' forearm skin, inform and establish a robust understanding of their complex hand activities. Fuelled by a neural-network-based model, our interface can decode finger-wise forces in real-time with 3.3% mean error, and generalize to new users with little calibration. Through an interactive psychophysical study, we show that human perception of virtual objects' physical properties, such as stiffness, can be significantly enhanced by our interface. We further demonstrate that our interface enables ubiquitous control via finger tapping. Ultimately, we envision our findings to push forward research towards more realistic physicality in future VR/AR., Comment: ACM Transactions on Graphics (SIGGRAPH Asia 2022)

Non-motor symptoms in Parkinson’s Disease (PD) predate motor symptoms and substantially decrease quality of life; however, detection, monitoring, and treatments are unavailable for many of these symptoms. Temporal perception abnormalities in PD are generally attributed to altered Basal Ganglia (BG) function. Present studies are confounded by motor control facilitating movements that are integrated into protocols assessing temporal perception. There is uncertainty regarding the BG’s influence on timing processes of different time scales and how PD therapies affect this perception. In this study, PD patients using Levodopa (n = 25), Deep Brain Stimulation (DBS; n = 6), de novo patients (n = 6), and healthy controls (n = 17) completed a visual temporal perception task in seconds and sub-section timing scales using a computer-generated graphical tool. For all patient groups, there were no impairments seen at the smaller tested magnitudes (using sub-second timing). However, all PD groups displayed significant impairments at the larger tested magnitudes (using interval timing). Neither Levodopa nor DBS therapy led to significant improvements in timing abilities. Levodopa resulted in a strong trend towards impairing timing processes and caused a deterioration in perceptual coherency according to Weber’s Law. It is shown that timing abnormalities in PD occur in the seconds range but do not extend to the sub-second range. Furthermore, observed timing deficits were shown to not be solely caused by motor deficiency. This provides evidence to support internal clock models involving the BG (among other neural regions) in interval timing, and cerebellar control of sub-second timing. This study also revealed significant temporal perception deficits in recently diagnosed PD patients; thus, temporal perception abnormalities might act as an early disease marker, with the graphical tool showing potential for disease monitoring.

The rapid development of the Internet and various applications such as the Internet of Medical Things (IoMT) has raised substantial concerns about personal information security. Conventional methods (e.g., passwords) and classic biological features (e.g., fingerprints) are security deficient because of potential information leakage and hacking. Biometrics that expressbehavioral featuressuggest a robust approach to achieving information security because of the corresponding uniqueness and complexity. In this paper, we consider identifying human subjects based on their transient neurophysiological signature captured using multichannel upper-limb surface electromyography (sEMG). An explainable artificial intelligence (XAI) approach is proposed to process the internal dynamics of temporal sEMG signals. We propose and prove the suitability of “transient sEMG” as a biomarker that can identify individuals. For this, we utilize the Gradient-weighted Class Activation Mapping (Grad-CAM) analysis to explain the network’s attention. The outcome not only decodes and makes the unique neurophysiological pattern (i.e., motor unit recruitment during the transient phase of contraction) associated with each individual visualizable but also generates an optimizing two-dimensional (2D) spectrotemporal mask used to significantly reduce the size of the model and the trainable parameters. The resulting mask selectively and systematically samples the spectrotemporal characteristics of the users’ neurophysiological responses, discarding 40% of the input space while securing the accuracy of about 74% with much shallower neural network architecture. In the systematic comparative study, we find that our proposed model outperforms several state-of-the-art algorithms. For broader impacts, we anticipate our design of a compact, practical, interpretable, and robust identification system that requires only a minimal number of gestures and sensors (only 7% of the entire data set) to be a starting point for small and portable identification hardware.

International audience; The advent of telerobotic systems has revolutionized various aspects of the industry and human life. This technology is designed to augment human sensorimotor capabilities to extend them beyond natural competence. Classic examples are space and underwater applications when distance and access are the two major physical barriers to be combated with this technology. In modern examples, telerobotic systems have been used in several clinical applications, including teleoperated surgery and telerehabilitation. In this regard, there has been a significant amount of research and development due to the major benefits in terms of medical outcomes. Recently telerobotic systems are combined with advanced artificial intelligence modules to better share the agency with the operator and open new doors of medical automation. In this review paper, we have provided a comprehensive analysis of the literature considering various topologies of telerobotic systems in the medical domain while shedding light on different levels of autonomy for this technology, starting from direct control, going up to command-tracking autonomous telerobots. Existing challenges, including instrumentation, transparency, autonomy, stochastic communication delays, and stability, in addition to the current direction of research related to benefit in telemedicine and medical automation, and future vision of this technology, are discussed in this review paper

Depression diagnosis is a challenging clinical task currently conducted mostly using subjective criteria. It is well known that depression alters the neural activity in the brain, so that the corresponding neurophysiological signature may be measured using non-invasive electroencephalography (EEG) signals. These, in turn, may be possible to decode using machine learning algorithms. Despite the extensive literature, the existing techniques rely on several channels and obtrusive systems. In this paper, and for the first time, the diagnostic power of each EEG channel for depression detection is analyzed using Neighborhood Component Analysis (NCA). Our results indicate that a mere two features collected from one EEG channel suffice for reliable diagnosis. To evaluate the performance of the proposed method, a dataset comprising seven minutes of EEG recording from 84 subjects is used. The data was divided into two separate sets, one for feature selection and one for diagnostic classification. We delineate brain regions that have the strongest discriminative power linked to depression diagnosis. Thus, we identified one electrode (i.e., AF4) located on the frontal lobe, which can be used to diagnose depression with high accuracy. After evaluation of a series of shallow machine learning methods, we achieved the classification accuracy of 80.8%, sensitivity of 60% and specificity of 99.7% with two features from one electrode. We also achieved the highest classification accuracy of 91.8%, the specificity of 93.5%, and sensitivity of 90% with two electrodes and three features. Our findings show that it is possible to significantly reduce the complexity of algorithms to diagnose depression with the motivation of use in highly accessible wearable devices.

Advances in biosignal signal processing and machine learning, in particular Deep Neural Networks (DNNs), have paved the way for the development of innovative Human-Machine Interfaces for decoding the human intent and controlling artificial limbs. DNN models have shown promising results with respect to other algorithms for decoding muscle electrical activity, especially for recognition of hand gestures. Such data-driven models, however, have been challenged by their need for a large number of trainable parameters and their structural complexity. Here we propose the novel Temporal Convolutions-based Hand Gesture Recognition architecture (TC-HGR) to reduce this computational burden. With this approach, we classified 17 hand gestures via surface Electromyogram (sEMG) signals by the adoption of attention mechanisms and temporal convolutions. The proposed method led to 81.65% and 80.72% classification accuracy for window sizes of 300ms and 200ms, respectively. The number of parameters to train the proposed TC-HGR architecture is 11.9 times less than that of its state-of-the-art counterpart.