SVM-Based Classification of sEMG Signals for Upper-Limb Self-Rehabilitation Training

Ternary spike-based neuromorphic signal processing system

Deep Neural Networks (DNNs) have been successfully implemented across various signal processing fields, resulting in significant enhancements in performance. However, DNNs generally require substantial computational resources, leading to significant economic costs and posing challenges for their deployment on resource-constrained edge devices. In this study, we take advantage of spiking neural networks (SNNs) and quantization technologies to develop an energy-efficient and lightweight neuromorphic signal processing system. Our system is characterized by two principal innovations: a threshold-adaptive encoding (TAE) method and a quantized ternary SNN (QT-SNN). The TAE method can efficiently encode time-varying analog signals into sparse ternary spike trains, thereby reducing energy and memory demands for signal processing. QT-SNN, compatible with ternary spike trains from the TAE method, quantifies both membrane potentials and synaptic weights to reduce memory requirements while maintaining performance. Extensive experiments are conducted on two typical signal-processing tasks: speech and electroencephalogram recognition. The results demonstrate that our neuromorphic signal processing system achieves state-of-the-art (SOTA) performance with a 94% reduced memory requirement. Furthermore, through theoretical energy consumption analysis, our system shows 7.5× energy saving compared to other SNN works. The efficiency and efficacy of the proposed system highlight its potential as a promising avenue for energy-efficient signal processing.

Systematic review of AI/ML applications in multi-domain robotic rehabilitation: trends, gaps, and future directions

Robotic technology is expected to transform rehabilitation settings, by providing precise, repetitive, and task-specific interventions, thereby potentially improving patients' clinical outcomes. Artificial intelligence (AI) and machine learning (ML) have been widely applied in different areas to support robotic rehabilitation, from controlling robot movements to real-time patient assessment. To provide an overview of the current landscape and the impact of AI/ML use in robotics rehabilitation, we performed a systematic review focusing on the use of AI and robotics in rehabilitation from a broad perspective, encompassing different pathologies and body districts, and considering both motor and neurocognitive rehabilitation. We searched the Scopus and IEEE Xplore databases, focusing on the studies involving human participants. After article retrieval, a tagging phase was carried out to devise a comprehensive and easily-interpretable taxonomy: its categories include the aim of the AI/ML within the rehabilitation system, the type of algorithms used, and the location of robots and sensors. The 201 selected articles span multiple domains and diverse aims, such as movement classification, trajectory prediction, and patient evaluation, demonstrating the potential of ML to revolutionize personalized therapy and improve patient engagement. ML is reported as highly effective in predicting movement intentions, assessing clinical outcomes, and detecting compensatory movements, providing insights into the future of personalized rehabilitation interventions. Our analysis also reveals pitfalls in the current use of AI/ML in this area, such as potential explainability issues and poor generalization ability when these systems are applied in real-world settings.

Design and Testing of a Portable Wireless Multi-Node sEMG System for Synchronous Muscle Signal Acquisition and Gesture Recognition

Surface electromyography (sEMG) is an important non-invasive method used in muscle function assessment, rehabilitation and human-machine interaction. However, existing commercial devices often lack sufficient channels, making it challenging to simultaneously acquire signals from multiple muscle sites.In this acticle, we design a portable multi-node sEMG acquisition system based on the TCP protocol to overcome the channel limitations of commercial sEMG detection devices. The system employs the STM32L442KCU6 microcontroller as the main control unit, with onboard ADC for analog-to-digital conversion of sEMG signals. Data filtered by analogy filter is transmitted via an ESP8266 WiFi module to the host computer for display and storage. By configuring Bluetooth broadcasting channels, the system can support up to 40 sEMG detection nodes. A gesture recognition algorithm is implemented to identify grasping motions with varying channel configurations. Experimental results demonstrate that with two channels, the Gradient Boosting Decision Tree (GBDT) algorithm achieves a recognition accuracy of 99.4%, effectively detecting grasping motions.

An efficient surface electromyography-based gesture recognition algorithm based on multiscale fusion convolution and channel attention

In the field of rehabilitation, although deep learning have been widely used in multitype gesture recognition via surface electromyography (sEMG), their higher algorithmic complexity often leads to low computationally inefficient, which compromise their practicality. To achieve more efficient multitype recognition, We propose the Residual-Inception-Efficient (RIE) model, which integrates Inception and efficient channel attention (ECA). The Inception, which is a multiscale fusion convolutional module, is adopted to enhance the ability to extract sEMG features. It uses fast dimensionality reduction, asymmetric convolution decomposition, and pooling to suppress the accumulation of parameters, then reducing the algorithmic complexity; The ECA is adopted to reweight the output features of Inception in different channels, enabling the RIE model to focus on information that is more relevant to gestures. 52-, 49-, and 52-class gesture recognition experiments are conducted on NinaPro DB1, DB3, and DB4 datasets, which contain 14,040, 3234, and 3120 gesture samples, respectively. RIE model proposed in this study achieves accuracies of 88.27%, 69.52%, and 84.55% on the three datasets, exhibiting excellent recognition accuracy and strong generalization ability. Moreover, this method reduces the algorithmic complexity from both spatial and temporal aspects, rendering it smaller in size and faster in computation compared to other lightweight algorithms. Therefore, the proposed RIE model possesses both lightweight computational requirements and reliable performance, providing an efficient deep learning method for gesture recognition based on sEMG.

AI Applications in Adult Stroke Recovery and Rehabilitation: A Scoping Review Using AI

Stroke is a leading cause of long-term disability worldwide. With the advancements in sensor technologies and data availability, artificial intelligence (AI) holds the promise of improving the amount, quality and efficiency of care and enhancing the precision of stroke rehabilitation. We aimed to identify and characterize the existing research on AI applications in stroke recovery and rehabilitation of adults, including categories of application and progression of technologies over time. Data were collected from peer-reviewed articles across various electronic databases up to January 2024. Insights were extracted using AI-enhanced multi-method, data-driven techniques, including clustering of themes and topics. This scoping review summarizes outcomes from 704 studies. Four common themes (impairment, assisted intervention, prediction and imaging, and neuroscience) were identified, in which time-linked patterns emerged. The impairment theme revealed a focus on motor function, gait and mobility, while the assisted intervention theme included applications of robotic and brain-computer interface (BCI) techniques. AI applications progressed over time, starting from conceptualization and then expanding to a broader range of techniques in supervised learning, artificial neural networks (ANN), natural language processing (NLP) and more. Applications focused on upper limb rehabilitation were reviewed in more detail, with machine learning (ML), deep learning techniques and sensors such as inertial measurement units (IMU) used for upper limb and functional movement analysis. AI applications have potential to facilitate tailored therapeutic delivery, thereby contributing to the optimization of rehabilitation outcomes and promoting sustained recovery from rehabilitation to real-world settings.

Advancements in Textile-Based sEMG Sensors for Muscle Fatigue Detection: A Journey from Material Evolution to Technological Integration

Textile-based surface electromyography (sEMG) electrodes have emerged as a prominent tool in muscle fatigue assessment, marking a significant shift toward innovative, noninvasive methods. This review examines the transition from metallic fibers to novel conductive polymers, elastomers, and advanced material-based electrodes, reflecting on the rapid evolution of materials in sEMG sensor technology. It highlights the pivotal role of materials science in enhancing sensor adaptability, signal accuracy, and longevity, crucial for practical applications in health monitoring, while examining the balance of clinical precision with user comfort. Additionally, it maps the global sEMG research landscape of diverse regional contributors and their impact on technological progress, focusing on the integration of Eastern manufacturing prowess with Western technological innovations and exploring both the opportunities and challenges in this global synergy. The integration of such textile-based sEMG innovations with artificial intelligence, nanotechnology, energy harvesting, and IoT connectivity is also anticipated as future prospects. Such advancements are poised to revolutionize personalized preventive healthcare. As the exploration of textile-based sEMG electrodes continues, the transformative potential not only promises to revolutionize integrated wellness and preventive healthcare but also signifies a seamless transition from laboratory innovations to real-world applications in sports medicine, envisioning the future of truly wearable material technologies.

A Novel TCN-LSTM Hybrid Model for sEMG-Based Continuous Estimation of Wrist Joint Angles

Surface electromyography (sEMG) offers a novel method in human-machine interactions (HMIs) since it is a distinct physiological electrical signal that conceals human movement intention and muscle information. Unfortunately, the nonlinear and non-smooth features of sEMG signals often make joint angle estimation difficult. This paper proposes a joint angle prediction model for the continuous estimation of wrist motion angle changes based on sEMG signals. The proposed model combines a temporal convolutional network (TCN) with a long short-term memory (LSTM) network, where the TCN can sense local information and mine the deeper information of the sEMG signals, while LSTM, with its excellent temporal memory capability, can make up for the lack of the ability of the TCN to capture the long-term dependence of the sEMG signals, resulting in a better prediction. We validated the proposed method in the publicly available Ninapro DB1 dataset by selecting the first eight subjects and picking three types of wrist-dependent movements: wrist flexion (WF), wrist ulnar deviation (WUD), and wrist extension and closed hand (WECH). Finally, the proposed TCN-LSTM model was compared with the TCN and LSTM models. The proposed TCN-LSTM outperformed the TCN and LSTM models in terms of the root mean square error (RMSE) and average coefficient of determination (R2). The TCN-LSTM model achieved an average RMSE of 0.064, representing a 41% reduction compared to the TCN model and a 52% reduction compared to the LSTM model. The TCN-LSTM also achieved an average R2 of 0.93, indicating an 11% improvement over the TCN model and an 18% improvement over the LSTM model.

Data-Driven Identification of Stroke through Machine Learning Applied to Complexity Metrics in Multimodal Electromyography and Kinematics

A stroke represents a significant medical condition characterized by the sudden interruption of blood flow to the brain, leading to cellular damage or death. The impact of stroke on individuals can vary from mild impairments to severe disability. Treatment for stroke often focuses on gait rehabilitation. Notably, assessing muscle activation and kinematics patterns using electromyography (EMG) and stereophotogrammetry, respectively, during walking can provide information regarding pathological gait conditions. The concurrent measurement of EMG and kinematics can help in understanding disfunction in the contribution of specific muscles to different phases of gait. To this aim, complexity metrics (e.g., sample entropy; approximate entropy; spectral entropy) applied to EMG and kinematics have been demonstrated to be effective in identifying abnormal conditions. Moreover, the conditional entropy between EMG and kinematics can identify the relationship between gait data and muscle activation patterns. This study aims to utilize several machine learning classifiers to distinguish individuals with stroke from healthy controls based on kinematics and EMG complexity measures. The cubic support vector machine applied to EMG metrics delivered the best classification results reaching 99.85% of accuracy. This method could assist clinicians in monitoring the recovery of motor impairments for stroke patients.

Post-stroke hand gesture recognition via one-shot transfer learning using prototypical networks

BACKGROUND

In-home rehabilitation systems are a promising, potential alternative to conventional therapy for stroke survivors. Unfortunately, physiological differences between participants and sensor displacement in wearable sensors pose a significant challenge to classifier performance, particularly for people with stroke who may encounter difficulties repeatedly performing trials. This makes it challenging to create reliable in-home rehabilitation systems that can accurately classify gestures.

METHODS

Twenty individuals who suffered a stroke performed seven different gestures (mass flexion, mass extension, wrist volar flexion, wrist dorsiflexion, forearm pronation, forearm supination, and rest) related to activities of daily living. They performed these gestures while wearing EMG sensors on the forearm, as well as FMG sensors and an IMU on the wrist. We developed a model based on prototypical networks for one-shot transfer learning, K-Best feature selection, and increased window size to improve model accuracy. Our model was evaluated against conventional transfer learning with neural networks, as well as subject-dependent and subject-independent classifiers: neural networks, LGBM, LDA, and SVM.

RESULTS

Our proposed model achieved 82.2% hand-gesture classification accuracy, which was better (P<0.05) than one-shot transfer learning with neural networks (63.17%), neural networks (59.72%), LGBM (65.09%), LDA (63.35%), and SVM (54.5%). In addition, our model performed similarly to subject-dependent classifiers, slightly lower than SVM (83.84%) but higher than neural networks (81.62%), LGBM (80.79%), and LDA (74.89%). Using K-Best features improved the accuracy in 3 of the 6 classifiers used for evaluation, while not affecting the accuracy in the other classifiers. Increasing the window size improved the accuracy of all the classifiers by an average of 4.28%.

CONCLUSION

Our proposed model showed significant improvements in hand-gesture recognition accuracy in individuals who have had a stroke as compared with conventional transfer learning, neural networks and traditional machine learning approaches. In addition, K-Best feature selection and increased window size can further improve the accuracy. This approach could help to alleviate the impact of physiological differences and create a subject-independent model for stroke survivors that improves the classification accuracy of wearable sensors.

TRIAL REGISTRATION NUMBER

The study was registered in Chinese Clinical Trial Registry with registration number CHiCTR1800017568 in 2018/08/04.

S-WD-EEMD: A hybrid framework for imbalanced sEMG signal analysis in diagnosis of human knee abnormality

The diagnosis of human knee abnormalities using the surface electromyography (sEMG) signal obtained from lower limb muscles with machine learning is a major problem due to the noisy nature of the sEMG signal and the imbalance in data corresponding to healthy and knee abnormal subjects. To address this challenge, a combination of wavelet decomposition (WD) with ensemble empirical mode decomposition (EEMD) and the Synthetic Minority Oversampling Technique (S-WD-EEMD) is proposed. In this study, a hybrid WD-EEMD is considered for the minimization of noises produced in the sEMG signal during the collection, while the Synthetic Minority Oversampling Technique (SMOTE) is considered to balance the data by increasing the minority class samples during the training of machine learning techniques. The findings indicate that the hybrid WD-EEMD with SMOTE oversampling technique enhances the efficacy of the examined classifiers when employed on the imbalanced sEMG data. The F-Score of the Extra Tree Classifier, when utilizing WD-EEMD signal processing with SMOTE oversampling, is 98.4%, whereas, without the SMOTE oversampling technique, it is 95.1%.

Stretchable, Flexible, Breathable, Self-Adhesive Epidermal Hand sEMG Sensor System

Hand function rehabilitation training typically requires monitoring the activation status of muscles directly related to hand function. However, due to factors such as the small surface area for hand-back electrode placement and significant skin deformation, the continuous real-time monitoring of high-quality surface electromyographic (sEMG) signals on the hand-back skin still poses significant challenges. We report a stretchable, flexible, breathable, and self-adhesive epidermal sEMG sensor system. The optimized serpentine structure exhibits a sufficient stretchability and filling ratio, enabling the high-quality monitoring of signals. The carving design minimizes the distribution of connecting wires, providing more space for electrode reservation. The low-cost fabrication design, combined with the cauterization design, facilitates large-scale production. Integrated with customized wireless data acquisition hardware, it demonstrates the real-time multi-channel sEMG monitoring capability for muscle activation during hand function rehabilitation actions. The sensor provides a new tool for monitoring hand function rehabilitation treatments, assessing rehabilitation outcomes, and researching areas such as prosthetic control.

Electrophysiological analysis of ENG signals in patients with Covid-19

Background

Currently, there is an increasing number of patients reporting dizziness, which has resulted in a positive COVID-19 PCR test. In this paper, we analyzed the ENG signals recorded from patients with a positive COVID-19 PCR test.

Methods

In this paper, both linear and nonlinear analyses of time series were employed to determine the regularity and complexity of a recorded ENG signal.

Results

The Wilcoxon rank-sum test indicated that the COVID-19 and non-COVID groups have significant differences based on different extracted features. Various machine learning methods including Linear Discriminant Analysis (LDA), Naïve Base (NB), K-nearest Neighbours (KNN), and Support Vector Machines (SVM) were used to classify COVID-19 and non-COVID groups. The best accuracy, precision and FCR achieved by SVM are 86%, 91% and 0.13.

Conclusion

In this study, ENG signals were recorded from COVID-19 and control groups. Linear and non-linear features were extracted from the recorded signals to identify significantly different features. Subjects were classified based on SVM and different classifiers. The SVM (polynomial kernel) classifier showed the best result. The proposed method had not been used for the classification of COVID-19 and non-COVID-19 subjects before. This work helps other researchers conduct more research on the development of machine learning methods to diagnose the COVID-19 virus using ENG and other physiological signals.

Home-Based Rehabilitation of the Shoulder Using Auxiliary Systems and Artificial Intelligence: An Overview

Advancements in modern medicine have bolstered the usage of home-based rehabilitation services for patients, particularly those recovering from diseases or conditions that necessitate a structured rehabilitation process. Understanding the technological factors that can influence the efficacy of home-based rehabilitation is crucial for optimizing patient outcomes. As technologies continue to evolve rapidly, it is imperative to document the current state of the art and elucidate the key features of the hardware and software employed in these rehabilitation systems. This narrative review aims to provide a summary of the modern technological trends and advancements in home-based shoulder rehabilitation scenarios. It specifically focuses on wearable devices, robots, exoskeletons, machine learning, virtual and augmented reality, and serious games. Through an in-depth analysis of existing literature and research, this review presents the state of the art in home-based rehabilitation systems, highlighting their strengths and limitations. Furthermore, this review proposes hypotheses and potential directions for future upgrades and enhancements in these technologies. By exploring the integration of these technologies into home-based rehabilitation, this review aims to shed light on the current landscape and offer insights into the future possibilities for improving patient outcomes and optimizing the effectiveness of home-based rehabilitation programs.

Research on control strategy of vehicle stability based on dynamic stable region regression analysis

The intervention time of stability control system is determined by stability judgment, which is the basis of vehicle stability control. According to the different working conditions of the vehicle, we construct the phase plane of the vehicle's sideslip angle and sideslip angular velocity, and establish the sample dataset of the stable region of the different phase planes. To reduce the complexity of phase plane stable region division and avoid large amount of data, we established the support vector regression (SVR) model, and realized the automatic regression of dynamic stable region. The testing of the test set shows that the model established in this paper has strong generalization ability. We designed a direct yaw-moment control (DYC) stability controller based on linear time-varying model predictive control (LTV-MPC). The influence of key factors such as centroid position and road adhesion coefficient on the stable region is analyzed through phase diagram. The effectiveness of the stability judgment and control algorithm is verified by simulation tests.

A novel attention-guided ECA-CNN architecture for sEMG-based gait classification

Gait recognition and classification technology is one of the essential technologies for detecting neurodegenerative dysfunction. This paper presents a gait classification model based on a convolutional neural network (CNN) with an efficient channel attention (ECA) module for gait detection applications using surface electromyographic (sEMG) signals. First, the sEMG sensor was used to collect the experimental sample data, and various gaits of different persons were collected to construct the sEMG signal data sets of different gaits. The CNN is used to extract the features of the one-dimensional input sEMG signal to obtain the feature vector, which is input into the ECA module to realize cross-channel interaction. Then, the next part of the convolutional layer is input to learn the signal features further. Finally, the model is output and tested to obtain the results. Comparative experiments show that the accuracy of the ECA-CNN network model can reach 97.75%.

A Clinical Perspective on Bespoke Sensing Mechanisms for Remote Monitoring and Rehabilitation of Neurological Diseases: Scoping Review

Neurological diseases including stroke and neurodegenerative disorders cause a hefty burden on the healthcare system. Survivors experience significant impairment in mobility and daily activities, which requires extensive rehabilitative interventions to assist them to regain lost skills and restore independence. The advent of remote rehabilitation architecture and enabling technology mandates the elaboration of sensing mechanisms tailored to individual clinical needs. This study aims to review current trends in the application of sensing mechanisms in remote monitoring and rehabilitation in neurological diseases, and to provide clinical insights to develop bespoke sensing mechanisms. A systematic search was performed using the PubMED database to identify 16 papers published for the period between 2018 to 2022. Teleceptive sensors (56%) were utilized more often than wearable proximate sensors (50%). The most commonly used modality was infrared (38%) and acceleration force (38%), followed by RGB color, EMG, light and temperature, and radio signal. The strategy adopted to improve the sensing mechanism included a multimodal sensor, the application of multiple sensors, sensor fusion, and machine learning. Most of the stroke studies utilized biofeedback control systems (78%) while the majority of studies for neurodegenerative disorders used sensors for remote monitoring (57%). Functional assessment tools that the sensing mechanism may emulate to produce clinically valid information were proposed and factors affecting user adoption were described. Lastly, the limitations and directions for further development were discussed.

Assistance level quantification-based human-robot interaction space reshaping for rehabilitation training

Stroke has become a major disease that seriously threatens human health due to its high incidence and disability rates. Most patients undergo upper limb motor dysfunction after stroke, which significantly impairs the ability of stroke survivors in their activities of daily living (ADL). Robots provide an optional solution for stroke rehabilitation by attending therapy in the hospital and the community, however, the rehabilitation robot still has difficulty in providing needed assistance interactively like human clinicians in conventional therapy. For safe and rehabilitation training, a human-robot interaction space reshaping method was proposed based on the recovery states of patients. According to different recovery states, we designed seven experimental protocols suitable for distinguishing rehabilitation training sessions. To achieve assist-as-needed (AAN) control, a PSO-SVM classification model and an LSTM-KF regression model were introduced to recognize the motor ability of patients with electromyography (EMG) and kinematic data, and a region controller for interaction space shaping was studied. Ten groups of offline and online experiments and corresponding data processing were conducted, and the machine learning and AAN control results were presented, which ensured the effective and the safe upper limb rehabilitation training. To discuss the human-robot interaction in different training stages and sessions, we defined a quantified assistance level index that characterizes the rehabilitation needs by considering the engagement of the patients and had the potential to apply in clinical upper limb rehabilitation training.

Joint mechanical properties estimation with a novel EMG-based knee rehabilitation robot: A machine learning approach

Joint dynamic properties play essential roles in a wide range of biomechanical movement control. This paper develops a device with a novel mechatronic design to apply small-amplitude perturbations to the human knee. Surface Electromyography is employed to record such information; at the same time, force and position sensors collect measurements to be sent to identify human joint dynamics. For classification and estimation of force, support vector machine and support vector regression techniques are applied, respectively. We devise a genetic algorithm for parameter optimization and feature extraction within the proposed methods to improve the estimation accuracy. These are then analyzed and compared to the output of our estimation model to provide a reliable comparison. Our extensive experimental results reveal a high estimation accuracy for lower limb muscles to regulate robot impedance parameters. Although the identification method sounds similar to traditional ones, knee joint properties can be estimated by the machine learning approach from the surface Electromyography without perturbations.

A Novel Method for Hand Movement Recognition Based on Wavelet Packet Transform and Principal Component Analysis with Surface Electromyogram

As an input method of signal language, the hand movement classification technology has developed into one of the ways of natural human-computer interaction. The surface electromyogram (sEMG) signal contains abundant human movement information and has significant advantages as the input signal of human-computer interaction. However, how to effectively extract components from sEMG signals to improve the accuracy of hand motion classification is a difficult problem. Therefore, this work proposes a novel method based on wavelet packet transform (WPT) and principal component analysis (PCA) to classify six kinds of hand motions. The method applies WPT to decompose the sEMG signal into multiple sub-band signals. To efficiently extract the intrinsic components of the sEMG signal, the classification performance of different wavelet packet basis functions is evaluated. The PCA algorithm is used to reduce the dimension of the feature space composed of the features reflecting hand motions extracted from each sub-band signal. Besides, to ensure higher classification performance while reducing the dimension of the feature space by the PCA algorithm, the classification performance of different dimensions of the feature space is compared. In addition, the effects of the variability of the sEMG signal and the size of the window on the proposed method are further analyzed. The proposed method was tested on the sEMG for Basic Hand Movements Data Set and achieved an average accuracy of 96.03%. Compared with the existing research, the proposed method has better classification performance, which indicates that the research results can be applied to the fields of exoskeleton robot, rehabilitation training, and intelligent prosthesis.

Proposal of a Wearable Multimodal Sensing-Based Serious Games Approach for Hand Movement Training After Stroke

Stroke often leads to hand motor dysfunction, and effective rehabilitation requires keeping patients engaged and motivated. Among the existing automated rehabilitation approaches, data glove-based systems are not easy to wear for patients due to spasticity, and single sensor-based approaches generally provided prohibitively limited information. We thus propose a wearable multimodal serious games approach for hand movement training after stroke. A force myography (FMG), electromyography (EMG), and inertial measurement unit (IMU)-based multi-sensor fusion model was proposed for hand movement classification, which was worn on the user’s affected arm. Two movement recognition-based serious games were developed for hand movement and cognition training. Ten stroke patients with mild to moderate motor impairments (Brunnstrom Stage for Hand II-VI) performed experiments while playing interactive serious games requiring 12 activities-of-daily-living (ADLs) hand movements taken from the Fugl Meyer Assessment. Feasibility was evaluated by movement classification accuracy and qualitative patient questionnaires. The offline classification accuracy using combined FMG-EMG-IMU was 81.0% for the 12 movements, which was significantly higher than any single sensing modality; only EMG, only FMG, and only IMU were 69.6, 63.2, and 47.8%, respectively. Patients reported that they were more enthusiastic about hand movement training while playing the serious games as compared to conventional methods and strongly agreed that they subjectively felt that the proposed training could be beneficial for improving upper limb motor function. These results showed that multimodal-sensor fusion improved hand gesture classification accuracy for stroke patients and demonstrated the potential of this proposed approach to be used as upper limb movement training after stroke.

Multimodal Human-Exoskeleton Interface for Lower Limb Movement Prediction Through a Dense Co-Attention Symmetric Mechanism

A challenging task for the biological neural signal-based human-exoskeleton interface is to achieve accurate lower limb movement prediction of patients with hemiplegia in rehabilitation training scenarios. The human-exoskeleton interface based on single-modal biological signals such as electroencephalogram (EEG) is currently not mature in predicting movements, due to its unreliability. The multimodal human-exoskeleton interface is a very novel solution to this problem. This kind of interface normally combines the EEG signal with surface electromyography (sEMG) signal. However, their use for the lower limb movement prediction is still limited—the connection between sEMG and EEG signals and the deep feature fusion between them are ignored. In this article, a Dense con-attention mechanism-based Multimodal Enhance Fusion Network (DMEFNet) is proposed for predicting lower limb movement of patients with hemiplegia. The DMEFNet introduces the con-attention structure to extract the common attention between sEMG and EEG signal features. To verify the effectiveness of DMEFNet, an sEMG and EEG data acquisition experiment and an incomplete asynchronous data collection paradigm are designed. The experimental results show that DMEFNet has a good movement prediction performance in both within-subject and cross-subject situations, reaching an accuracy of 82.96 and 88.44%, respectively.

Dual-Modal Hybrid Control for an Upper-Limb Rehabilitation Robot

The recovery treatment of motor dysfunction plays a crucial role in rehabilitation therapy. Rehabilitation robots are partially or fully replacing therapists in assisting patients in exercise by advantage of robot technologies. However, the rehabilitation training system is not yet intelligent enough to provide suitable exercise modes based on the exercise intentions of patients with different motor abilities. In this paper, a dual-modal hybrid self-switching control strategy (DHSS) is proposed to automatically determine the exercise mode of patients, i.e., passive and assistive exercise mode. In this strategy, the potential field method and the ADRC position control are employed to plan trajectories and assist patients’ training. Dual-modal self-switching rules based on the motor and impulse information of patients are presented to identify patients’ motor abilities. Finally, the DHSS assisted five subjects in performing the training with an average deviation error of less than 2 mm in both exercise modes. The experimental results demonstrate that the muscle activation of the subjects differed significantly in different modes. It also verifies that DHSS is reasonable and effective, which helps patients to train independently without therapists.

Machine Learning for Detection of Muscular Activity from Surface EMG Signals

Background: Muscular-activity timing is useful information that is extractable from surface EMG signals (sEMG). However, a reference method is not available yet. The aim of this study is to investigate the reliability of a novel machine-learning-based approach (DEMANN) in detecting the onset/offset timing of muscle activation from sEMG signals. Methods: A dataset of 2880 simulated sEMG signals, stratified for signal-to-noise ratio (SNR) and time support, was generated to train a hidden single-layer fully-connected neural network. DEMANN’s performance was evaluated on simulated sEMG signals and two different datasets of real sEMG signals. DEMANN was validated against different reference algorithms, including the acknowledged double-threshold statistical algorithm (DT). Results: DEMANN provided a reliable prediction of muscle onset/offset in simulated and real sEMG signals, being minimally affected by SNR variability. When directly compared with state-of-the-art algorithms, DEMANN introduced relevant improvements in prediction performances. Conclusions: These outcomes support DEMANN’s reliability in assessing onset/offset events in different motor tasks and the condition of signal quality (different SNR), improving reference-algorithm performances. Unlike other works, DEMANN’s adopts a machine learning approach where a neural network is trained by only simulated sEMG signals, avoiding the possible complications and costs associated with a typical experimental procedure, making this approach suitable to clinical practice.

Comparative Analysis of Supervised Classifiers for the Evaluation of Sarcopenia Using a sEMG-Based Platform

Sarcopenia is a geriatric condition characterized by a loss of strength and muscle mass, with a high impact on health status, functional independence and quality of life in older adults. To reduce the effects of the disease, just the diagnostic is not enough, it is necessary more than recognition. Surface electromyography is becoming increasingly relevant for the prevention and diagnosis of sarcopenia, also due to a wide diffusion of smart and minimally invasive wearable devices suitable for electromyographic monitoring. The purpose of this work is manifold. The first aim is the design and implementation of a hardware/software platform. It is based on the elaboration of surface electromyographic signals extracted from the Gastrocnemius Lateralis and Tibialis Anterior muscles, useful to analyze the strength of the muscles with the purpose of distinguishing three different “confidence” levels of sarcopenia. The second aim is to compare the efficiency of state of the art supervised classifiers in the evaluation of sarcopenia. The experimentation stage was performed on an “augmented” dataset starting from data acquired from 32 patients. The latter were distributed in an unbalanced manner on 3 “confidence” levels of sarcopenia. The obtained results in terms of classification accuracy demonstrated the ability of the proposed platform to distinguish different sarcopenia “confidence” levels, with highest accuracy value given by Support Vector Machine classifier, outperforming the other classifiers by an average of 7.7%.

Stacking classifier to improve the classification of shoulder motion in transhumeral amputees

In recent years surface electromyography signals-based machine learning models are rapidly establishing. The efficacy of prosthetic arm growth for transhumeral amputees is aided by efficient classifiers. The paper aims to propose a stacking classifier-based classification system for sEMG shoulder movements. It presents the possibility of various shoulder motions classification of transhumeral amputees. To improve the system performance, adaptive threshold method and wavelet transformation have been applied for features extraction. Six different classifiers Support Vector Machines (SVM), Tree, Random Forest (RF), K-Nearest Neighbour (KNN), AdaBoost and Naïve Bayes (NB) are designed to extract the sEMG data classification accuracy. With cross-validation, the accuracy of RF, Tree and Ada Boost is 97%, 92% and 92% respectively. Stacking classifiers provides an accuracy as 99.4% after combining the best predicted multiple classifiers.

Muscle force estimation from lower limb EMG signals using novel optimised machine learning techniques

The main objective of this work is to establish a framework for processing and evaluating the lower limb electromyography (EMG) signals ready to be fed to a rehabilitation robot. We design and build a knee rehabilitation robot that works with surface EMG (sEMG) signals. In our device, the muscle forces are estimated from sEMG signals using several machine learning techniques, i.e. support vector machine (SVM), support vector regression (SVR) and random forest (RF). In order to improve the estimation accuracy, we devise genetic algorithm (GA) for parameter optimisation and feature extraction within the proposed methods. At the same time, a load cell and a wearable inertial measurement unit (IMU) are mounted on the robot to measure the muscle force and knee joint angle, respectively. Various performance measures have been employed to assess the performance of the proposed system. Our extensive experiments and comparison with related works revealed a high estimation accuracy of 98.67% for lower limb muscles. The main advantage of the proposed techniques is high estimation accuracy leading to improved performance of the therapy while muscle models become especially sensitive to the tendon stiffness and the slack length.

MCSNet: Channel Synergy-Based Human-Exoskeleton Interface With Surface Electromyogram

The human-robot interface (HRI) based on biological signals can realize the natural interaction between human and robot. It has been widely used in exoskeleton robots recently to help predict the wearer's movement. Surface electromyography (sEMG)-based HRI has mature applications on the exoskeleton. However, the sEMG signals of paraplegic patients' lower limbs are weak, which means that most HRI based on lower limb sEMG signals cannot be applied to the exoskeleton. Few studies have explored the possibility of using upper limb sEMG signals to predict lower limb movement. In addition, most HRIs do not consider the contribution and synergy of sEMG signal channels. This paper proposes a human-exoskeleton interface based on upper limb sEMG signals to predict lower limb movements of paraplegic patients. The interface constructs an channel synergy-based network (MCSNet) to extract the contribution and synergy of different feature channels. An sEMG data acquisition experiment is designed to verify the effectiveness of MCSNet. The experimental results show that our method has a good movement prediction performance in both within-subject and cross-subject situations, reaching an accuracy of 94.51 and 80.75%, respectively. Furthermore, feature visualization and model ablation analysis show that the features extracted by MCSNet are physiologically interpretable.

Voting-based 1D CNN model for human lower limb activity recognition using sEMG signal

Surface electromyography (sEMG) signal classification has many applications such as human-machine interaction, diagnosis of kinesiological studies, and neuromuscular diseases. However, these signals are complicated because of different artifacts added to the sEMG signal during recording. In this study, a multi-stage classification technique is proposed for the identification of distinct movements of the lower limbs using sEMG signals acquired from leg muscles of healthy knee and abnormal knee subjects. This investigation involves 11 subjects with a knee abnormality and 11 without knee abnormality for three distinct activities viz. walking, leg extension from sitting position (sitting), and flexion of the leg (standing). Discrete wavelet denoising to fourth level decomposition has been implemented for the artifact reduction and the signal has been segmented using overlapping windowing technique. A study of four different architectures of 1D convolutional neural network models is undertaken for the prediction of lower limb activities and the final prediction is achieved via a voting mechanism of all four model results. The performance parameters of CNN models have been calculated for three different cases: (1) healthy subjects (2) subjects with knee abnormality (3) Pooled data (combination of abnormal knee and healthy knee subjects) using nested threefold cross-validation. It has been found that the voting mechanism yields an average classification accuracy as 99.35%, 97.63%, and 97.14% for healthy subjects, knee abnormal subjects, and pooled data, respectively. The result validates that the proposed voting-based 1D CNN model is efficient and useful in lower limb activity recognition using the sEMG signal.

Self-Rehabilitation Guided by a Mobile Application After Anterior Cruciate Ligament Reconstruction Leads to Improved Early Motion and Less Pain

Purpose

To evaluate the adherence rate and the contribution of self-rehabilitation (SR) guided by a mobile application after anterior cruciate ligament reconstruction (ACLR) in combination with physical therapy sessions on early knee function.

Methods

This study was a retrospective analysis of prospectively collected data from a single health care facility. All patients who underwent ACLR by a single surgeon from December 2019 to September 2020 were included. Two groups were formed and compared based on use of the mobile app: users (>10 days of use) and nonusers (≤10 days of use). Outcomes included physical examination at 3 and 6 weeks postoperatively.

Results

A total of 65 patients were analyzed: 19 in the nonuser group and 46 in the user group. Adherence rate was 91% at 10 days, 71% at 15 days, 62% at 21 days, and 44% at 45 days. At 3 weeks, the user group was 3.86 times [range 1.12 to 13.3] as likely to lock the quadriceps during gait with crutches and was 4.2 times [range 1.2 to 14.3] as likely to be pain free. There was a tendency to have less flexion contracture in the user group (17% versus 32%, P = .32). At 6 weeks, the differences leveled out, but the user group still had slightly better quadriceps locking during gait without crutches (87% versus 79%, P = .46).

Conclusions

SR guided by a mobile app combined with a standard rehabilitation protocol is correlated with better knee function at initial follow-up.

Level of evidence

IV, therapeutic case series.

Inter-classifier comparison for upper extremity EMG signal at different hand postures and arm positions using pattern recognition

The utilization of surface EMG and intramuscular EMG signals has been observed to create significant improvement in pattern recognition approaches and myoelectric control. However, there is less data of different arm positions and hand postures available. Hand postures and arm positions tend to affect the combination of surface and intramuscular EMG signal acquisition in terms of classifier accuracy. Hence, this study aimed to find a robust classifier for two scenarios: (1) at fixed arm position (FAP) where classifiers classify different hand postures and (2) at fixed hand posture (FHP) where classifiers classify different arm positions. A total of 20 healthy male participants (30.62 ± 3.87 years old) were recruited for this study. They were asked to perform five motion classes including hand grasp, hand open, rest, hand extension, and hand flexion at four different arm positions at 0°, 45°, 90°, and 135°. SVM, KNN, and LDA classifier were deployed. Statistical analysis in the form of pairwise comparisons was carried out using SPSS. It is concluded that there is no significant difference among the three classifiers. SVM gave highest accuracy of 75.35% and 58.32% at FAP and FHP respectively for each motion classification. KNN yielded the highest accuracies of 69.11% and 79.04% when data was pooled and was classified at different arm positions and at different hand postures respectively. The results exhibited that there is no significant effect of changing arm position and hand posture on the classifier accuracy.

Biosignal-Based Human-Machine Interfaces for Assistance and Rehabilitation: A Survey

As a definition, Human–Machine Interface (HMI) enables a person to interact with a device. Starting from elementary equipment, the recent development of novel techniques and unobtrusive devices for biosignals monitoring paved the way for a new class of HMIs, which take such biosignals as inputs to control various applications. The current survey aims to review the large literature of the last two decades regarding biosignal-based HMIs for assistance and rehabilitation to outline state-of-the-art and identify emerging technologies and potential future research trends. PubMed and other databases were surveyed by using specific keywords. The found studies were further screened in three levels (title, abstract, full-text), and eventually, 144 journal papers and 37 conference papers were included. Four macrocategories were considered to classify the different biosignals used for HMI control: biopotential, muscle mechanical motion, body motion, and their combinations (hybrid systems). The HMIs were also classified according to their target application by considering six categories: prosthetic control, robotic control, virtual reality control, gesture recognition, communication, and smart environment control. An ever-growing number of publications has been observed over the last years. Most of the studies (about 67%) pertain to the assistive field, while 20% relate to rehabilitation and 13% to assistance and rehabilitation. A moderate increase can be observed in studies focusing on robotic control, prosthetic control, and gesture recognition in the last decade. In contrast, studies on the other targets experienced only a small increase. Biopotentials are no longer the leading control signals, and the use of muscle mechanical motion signals has experienced a considerable rise, especially in prosthetic control. Hybrid technologies are promising, as they could lead to higher performances. However, they also increase HMIs’ complexity, so their usefulness should be carefully evaluated for the specific application.

Upper-Limb Motion Recognition Based on Hybrid Feature Selection: Algorithm Development and Validation

Background

For rehabilitation training systems, it is essential to automatically record and recognize exercises, especially when more than one type of exercise is performed without a predefined sequence. Most motion recognition methods are based on feature engineering and machine learning algorithms. Time-domain and frequency-domain features are extracted from original time series data collected by sensor nodes. For high-dimensional data, feature selection plays an important role in improving the performance of motion recognition. Existing feature selection methods can be categorized into filter and wrapper methods. Wrapper methods usually achieve better performance than filter methods; however, in most cases, they are computationally intensive, and the feature subset obtained is usually optimized only for the specific learning algorithm.

Objective

This study aimed to provide a feature selection method for motion recognition of upper-limb exercises and improve the recognition performance.

Methods

Motion data from 5 types of upper-limb exercises performed by 21 participants were collected by a customized inertial measurement unit (IMU) node. A total of 60 time-domain and frequency-domain features were extracted from the original sensor data. A hybrid feature selection method by combining filter and wrapper methods (FESCOM) was proposed to eliminate irrelevant features for motion recognition of upper-limb exercises. In the filter stage, candidate features were first selected from the original feature set according to the significance for motion recognition. In the wrapper stage, k-nearest neighbors (kNN), Naïve Bayes (NB), and random forest (RF) were evaluated as the wrapping components to further refine the features from the candidate feature set. The performance of the proposed FESCOM method was verified using experiments on motion recognition of upper-limb exercises and compared with the traditional wrapper method.

Results

Using kNN, NB, and RF as the wrapping components, the classification error rates of the proposed FESCOM method were 1.7%, 8.9%, and 7.4%, respectively, and the feature selection time in each iteration was 13 seconds, 71 seconds, and 541 seconds, respectively.

Conclusions

The experimental results demonstrated that, in the case of 5 motion types performed by 21 healthy participants, the proposed FESCOM method using kNN and NB as the wrapping components achieved better recognition performance than the traditional wrapper method. The FESCOM method dramatically reduces the search time in the feature selection process. The results also demonstrated that the optimal number of features depends on the classifier. This approach serves to improve feature selection and classification algorithm selection for upper-limb motion recognition based on wearable sensor data, which can be extended to motion recognition of more motion types and participants.

Recognizing Physical Activities for Spinal Cord Injury Rehabilitation Using Wearable Sensors

The research area of activity recognition is fast growing with diverse applications. However, advances in this field have not yet been used to monitor the rehabilitation of individuals with spinal cord injury. Noteworthily, relying on patient surveys to assess adherence can undermine the outcomes of rehabilitation. Therefore, this paper presents and implements a systematic activity recognition method to recognize physical activities applied by subjects during rehabilitation for spinal cord injury. In the method, raw sensor data are divided into fragments using a dynamic segmentation technique, providing higher recognition performance compared to the sliding window, which is a commonly used approach. To develop the method and build a predictive model, a machine learning approach was adopted. The proposed method was evaluated on a dataset obtained from a single wrist-worn accelerometer. The results demonstrated the effectiveness of the proposed method in recognizing all of the activities that were examined, and it achieved an overall accuracy of 96.86%.

Dynamic Segmentation for Physical Activity Recognition Using a Single Wearable Sensor

Data segmentation is an essential process in activity recognition when using machine learning techniques. Previous studies on physical activity recognition have mostly relied on the sliding window approach for segmentation. However, choosing a fixed window size for multiple activities with different durations may affect recognition accuracy, especially when the activities belong to the same category (i.e., dynamic or static). This paper presents and verifies a new method for dynamic segmentation of physical activities performed during the rehabilitation of individuals with spinal cord injuries. To adaptively segment the raw data, signal characteristics are analyzed to determine the suitable type of boundaries. Then, the algorithm identifies the time boundaries to represent the start- and endpoints of each activity. To verify the method and build a predictive model, an experiment was conducted in which data were collected using a single wrist-worn accelerometer sensor. The experimental results were compared with the sliding window approach, indicating that the proposed method outperformed the sliding window approach in terms of overall accuracy, which exceeded 5%, as well as model robustness. The results also demonstrated efficient physical activity segmentation using the proposed method, resulting in high classification performance for all activities considered.

Minimum Mapping from EMG Signals at Human Elbow and Shoulder Movements into Two DoF Upper-Limb Robot with Machine Learning

This research focuses on the minimum process of classifying three upper arm movements (elbow extension, shoulder extension, combined shoulder and elbow extension) of humans with three electromyography (EMG) signals, to control a 2-degrees of freedom (DoF) robotic arm. The proposed minimum process consists of four parts: time divisions of data, Teager–Kaiser energy operator (TKEO), the conventional EMG feature extraction (i.e., the mean absolute value (MAV), zero crossings (ZC), slope-sign changes (SSC), and waveform length (WL)), and eight major machine learning models (i.e., decision tree (medium), decision tree (fine), k-Nearest Neighbor (KNN) (weighted KNN, KNN (fine), Support Vector Machine (SVM) (cubic and fine Gaussian SVM), Ensemble (bagged trees and subspace KNN). Then, we compare and investigate 48 classification models (i.e., 47 models are proposed, and 1 model is the conventional) based on five healthy subjects. The results showed that all the classification models achieved accuracies ranging between 74–98%, and the processing speed is below 40 ms and indicated acceptable controller delay for robotic arm control. Moreover, we confirmed that the classification model with no time division, with TKEO, and with ensemble (subspace KNN) had the best performance in accuracy rates at 96.67, recall rates at 99.66, and precision rates at 96.99. In short, the combination of the proposed TKEO and ensemble (subspace KNN) plays an important role to achieve the EMG classification.

Lw-CNN-Based Myoelectric Signal Recognition and Real-Time Control of Robotic Arm for Upper-Limb Rehabilitation

Deep-learning models can realize the feature extraction and advanced abstraction of raw myoelectric signals without necessitating manual selection. Raw surface myoelectric signals are processed with a deep model in this study to investigate the feasibility of recognizing upper-limb motion intents and real-time control of auxiliary equipment for upper-limb rehabilitation training. Surface myoelectric signals are collected on six motions of eight subjects' upper limbs. A light-weight convolutional neural network (Lw-CNN) and support vector machine (SVM) model are designed for myoelectric signal pattern recognition. The offline and online performance of the two models are then compared. The average accuracy is (90 ± 5)% for the Lw-CNN and (82.5 ± 3.5)% for the SVM in offline testing of all subjects, which prevails over (84 ± 6)% for the online Lw-CNN and (79 ± 4)% for SVM. The robotic arm control accuracy is (88.5 ± 5.5)%. Significance analysis shows no significant correlation (p = 0.056) among real-time control, offline testing, and online testing. The Lw-CNN model performs well in the recognition of upper-limb motion intents and can realize real-time control of a commercial robotic arm.

The realization of robotic neurorehabilitation in clinical: use of computational intelligence and future prospects analysis

Introduction: Although there is a need for rehabilitation treatment with the increase in the aging population, the shortage of skilled physicians frustrates this necessity. Robotic technology has been advocated as one of the most viable methods with the potential to replace humans in providing physical rehabilitation of patients with neurological impairment. However, because the pioneering robot devices suffer several reservations such as safety and comfort concerns in clinical practice, there is an urgent need to provide upgraded replacements. The rapid development of intelligent computing has attracted the attention of researchers concerning the utilization of computational intelligence algorithms for robots in rehabilitation. Areas covered: This article reviews the state of the art and advances of robotic neurorehabilitation with computational intelligence. We classified advances into two categories: mechanical structures and control methods. Prospective outlooks of rehabilitation robots also have been discussed. Expert opinion: The aggravation of global aging has promoted the application of robotic technology in neurorehabilitation. However, this approach is not mature enough to guarantee the safety of patients. Our critical review summarizes multiple computation algorithms which have been proved to be valuable for better robotic use in clinical settings and guide the possible future advances in this industry.

Shoulder muscle activation pattern recognition based on sEMG and machine learning algorithms

BACKGROUND AND OBJECTIVE

Surface electromyography (sEMG) has been used for robotic rehabilitation engineering for volitional control of hand prostheses or elbow exoskeleton, however, using sEMG for volitional control of an upper limb exoskeleton has not been perfectly developed. The long-term goal of our study is to process shoulder muscle bio-electrical signals for rehabilitative robotic assistive device motion control. The purposes of this study included: 1) to test the feasibility of machine learning algorithms in shoulder motion pattern recognition using sEMG signals from shoulder and upper limb muscles, 2) to investigate the influence of motion speed, individual variability, EMG recording device, and the amount of EMG datasets on the shoulder motion pattern recognition accuracy.

METHODS

A novel convolutional neural network (CNN) structure was constructed to process EMG signals from 12 muscles for the pattern recognition of upper arm motions including resting, drinking, backward-forward motion, and abduction motion. The accuracy of the CNN models for pattern recognition under different motion speeds, among individuals, and by EMG recording devices was statistically analyzed using ANOVA, GLM Univariate analysis, and Chi-square tests. The influence of EMG dataset number used for CNN model training on recognition accuracy was studied by gradually increasing dataset number until the highest accuracy was obtained.

RESULTS

Results showed that the accuracy of the normal speed CNN model in motion pattern recognition was 97.57% for normal speed motions and 97.07% for fast speed motions. The accuracy of the cross-subjects CNN model in motion pattern recognition was 79.64%. The accuracy of the cross-device CNN model in motion pattern recognition was 88.93% for normal speed motion and 80.87% for mixed speed. There was a statistical difference in pattern recognition accuracy between different CNN models.

CONCLUSION

The EMG signals of shoulder and upper arm muscles from the upper limb motions can be processed using CNN algorithms to recognize the identical motions of the upper limb including drinking, forward/backward, abduction, and resting. A simple CNN model trained by EMG datasets of a designated motion speed accurately detected the motion patterns of the same motion speed, yielding the highest accuracy compared with other mixed CNN models for various speeds of motion pattern recognition. Increase of the number of EMG datasets for CNN model training improved the pattern recognition accuracy.