Surface EMG signals and deep transfer learning-based physical action classification

Across Sessions and Subjects Domain Adaptation for Building Robust Myoelectric Interface

Gesture interaction via surface electromyography (sEMG) signal is a promising approach for advanced human-computer interaction systems. However, improving the performance of the myoelectric interface is challenging due to the domain shift caused by the signal's inherent variability. To enhance the interface's robustness, we propose a novel adaptive information fusion neural network (AIFNN) framework, which could effectively reduce the effects of multiple scenarios. Specifically, domain adversarial training is established to inhibit the shared network's weights from exploiting domain-specific representation, thus allowing for the extraction of domain-invariant features. Effectively, classification loss, domain diversence loss and domain discrimination loss are employed, which improve classification performance while reduce distribution mismatches between the two domains. To simulate the application of myoelectric interface, experiments were carried out involving three scenarios (intra-session, inter-session and inter-subject scenarios). Ten non-disabled subjects were recruited to perform sixteen gestures for ten consecutive days. The experimental results indicated that the performance of AIFNN was better than two other state-of-the-art transfer learning approaches, namely fine-tuning (FT) and domain adversarial network (DANN). This study demonstrates the capability of AIFNN to maintain robustness over time and generalize across users in practical myoelectric interface implementations. These findings could serve as a foundation for future deployments.

Classification of EMG signals with CNN features and voting ensemble classifier

Electromyography (EMG) signals are primarily used to control prosthetic hands. Classifying hand gestures efficiently with EMG signals presents numerous challenges. In addition to overcoming these challenges, a successful combination of feature extraction and classification approaches will improve classification accuracy. In the current work, convolutional neural network (CNN) features are used to reduce the redundancy problems associated with time and frequency domain features to improve classification accuracy. The features from the EMG signal are extracted using a CNN and are fed to the 'k' nearest neighbor (KNN) classifier with a different number of neighbors ( 1 N N , 3 N N , 5 N N , and   7 N N ) . It results in an ensemble of classifiers that are combined using a hard voting-based classifier. Based on the benchmark Ninapro DB4 database and CapgMyo database, the proposed framework obtained 91.3 % classification accuracy on CapgMyo and 89.5 % on Ninapro DB4.

Biohybrid Robotic Hand to Investigate Tactile Encoding and Sensorimotor Integration

For people who have experienced a spinal cord injury or an amputation, the recovery of sensation and motor control could be incomplete despite noteworthy advances with invasive neural interfaces. Our objective is to explore the feasibility of a novel biohybrid robotic hand model to investigate aspects of tactile sensation and sensorimotor integration with a pre-clinical research platform. Our new biohybrid model couples an artificial hand with biological neural networks (BNN) cultured in a multichannel microelectrode array (MEA). We decoded neural activity to control a finger of the artificial hand that was outfitted with a tactile sensor. The fingertip sensations were encoded into rapidly adapting (RA) or slowly adapting (SA) mechanoreceptor firing patterns that were used to electrically stimulate the BNN. We classified the coherence between afferent and efferent electrodes in the MEA with a convolutional neural network (CNN) using a transfer learning approach. The BNN exhibited the capacity for functional specialization with the RA and SA patterns, represented by significantly different robotic behavior of the biohybrid hand with respect to the tactile encoding method. Furthermore, the CNN was able to distinguish between RA and SA encoding methods with 97.84% ± 0.65% accuracy when the BNN was provided tactile feedback, averaged across three days in vitro (DIV). This novel biohybrid research platform demonstrates that BNNs are sensitive to tactile encoding methods and can integrate robotic tactile sensations with the motor control of an artificial hand. This opens the possibility of using biohybrid research platforms in the future to study aspects of neural interfaces with minimal human risk.

TraHGR: Transformer for Hand Gesture Recognition via Electromyography

Deep learning-based Hand Gesture Recognition (HGR) via surface Electromyogram (sEMG) signals have recently shown considerable potential for development of advanced myoelectric-controlled prosthesis. Although deep learning techniques can improve HGR accuracy compared to their classical counterparts, classifying hand movements based on sparse multichannel sEMG signals is still a challenging task. Furthermore, existing deep learning approaches, typically, include only one model as such can hardly extract representative features. In this paper, we aim to address this challenge by capitalizing on the recent advances in hybrid models and transformers. In other words, we propose a hybrid framework based on the transformer architecture, which is a relatively new and revolutionizing deep learning model. The proposed hybrid architecture, referred to as the Transformer for Hand Gesture Recognition (TraHGR), consists of two parallel paths followed by a linear layer that acts as a fusion center to integrate the advantage of each module. We evaluated the proposed architecture TraHGR based on the commonly used second Ninapro dataset, referred to as the DB2. The sEMG signals in the DB2 dataset are measured in real-life conditions from 40 healthy users, each performing 49 gestures. We have conducted an extensive set of experiments to test and validate the proposed TraHGR architecture, and compare its achievable accuracy with several recently proposed HGR classification algorithms over the same dataset. We have also compared the results of the proposed TraHGR architecture with each individual path and demonstrated the distinguishing power of the proposed hybrid architecture. The recognition accuracies of the proposed TraHGR architecture for the window of size 200ms and step size of 100ms are 86.00%, 88.72%, 81.27%, and 93.74%, which are 2.30%, 4.93%, 8.65%, and 4.20% higher than the state-of-the-art performance for DB2 (49 gestures), DB2-B (17 gestures), DB2-C (23 gestures), and DB2-D (9 gestures), respectively.

A Smartphone-Based sEMG Signal Analysis System for Human Action Recognition

In lower-limb rehabilitation, human action recognition (HAR) technology can be introduced to analyze the surface electromyography (sEMG) signal generated by movements, which can provide an objective and accurate evaluation of the patient's action. To balance the long cycle required for rehabilitation and the inconvenient factors brought by wearing sEMG devices, a portable sEMG signal acquisition device was developed that can be used under daily scenarios. Additionally, a mobile application was developed to meet the demand for real-time monitoring and analysis of sEMG signals. This application can monitor data in real time and has functions such as plotting, filtering, storage, and action capture and recognition. To build the dataset required for the recognition model, six lower-limb motions were developed for rehabilitation (kick, toe off, heel off, toe off and heel up, step back and kick, and full gait). The sEMG segment and action label were combined for training a convolutional neural network (CNN) to achieve high-precision recognition performance for human lower-limb actions (with a maximum accuracy of 97.96% and recognition accuracy for all actions reaching over 97%). The results show that the smartphone-based sEMG analysis system proposed in this paper can provide reliable information for the clinical evaluation of lower-limb rehabilitation.

Myoelectric Pattern Recognition Using Gramian Angular Field and Convolutional Neural Networks for Muscle-Computer Interface

In the field of the muscle-computer interface, the most challenging task is extracting patterns from complex surface electromyography (sEMG) signals to improve the performance of myoelectric pattern recognition. To address this problem, a two-stage architecture, consisting of Gramian angular field (GAF)-based 2D representation and convolutional neural network (CNN)-based classification (GAF-CNN), is proposed. To explore discriminant channel features from sEMG signals, sEMG-GAF transformation is proposed for time sequence signal representation and feature modeling, in which the instantaneous values of multichannel sEMG signals are encoded in image form. A deep CNN model is introduced to extract high-level semantic features lying in image-form-based time sequence signals concerning instantaneous values for image classification. An insight analysis explains the rationale behind the advantages of the proposed method. Extensive experiments are conducted on benchmark publicly available sEMG datasets, i.e., NinaPro and CagpMyo, whose experimental results validate that the proposed GAF-CNN method is comparable to the state-of-the-art methods, as reported by previous work incorporating CNN models.

Towards Instant Calibration in Myoelectric Prosthetic Hands: A Highly Data-Efficient Controller Based on the Wasserstein Distance

Prosthetic hand control research typically focuses on developing increasingly complex controllers to achieve diminishing returns in pattern recognition of muscle activity signals, making models less suitable for user calibration. Some works have investigated transfer learning to alleviate this, but such approaches increase model size dramatically-thus reducing their suitability for implementation on real prostheses. In this work, we propose a novel, non-parametric controller that uses the Wasserstein distance to compare the distribution of incoming signals to those of a set of reference distributions, with the intended action classified as the closest distribution. This controller requires only a single capture of data per reference distribution, making calibration almost instantaneous. Preliminary experiments building a reference library show that, in theory, users are able to produce up to 9 distinguishable muscle activity patterns. However, in practice, variation when repeating actions reduces this. Controller accuracy results show that 10 non-disabled and 1 disabled participant were able to use the controller with a maximum of two recalibrations to perform 6 actions at an average accuracy of 89.9% and 86.7% respectively. Practical experiments show that the controller allows users to complete all tasks of the Jebsen-Taylor Hand Function Test, although the task of picking and placing small common objects required on average more time than the protocol's maximum time.

A Systematic Study on Electromyography-Based Hand Gesture Recognition for Assistive Robots Using Deep Learning and Machine Learning Models

Upper limb amputation severely affects the quality of life and the activities of daily living of a person. In the last decade, many robotic hand prostheses have been developed which are controlled by using various sensing technologies such as artificial vision and tactile and surface electromyography (sEMG). If controlled properly, these prostheses can significantly improve the daily life of hand amputees by providing them with more autonomy in physical activities. However, despite the advancements in sensing technologies, as well as excellent mechanical capabilities of the prosthetic devices, their control is often limited and usually requires a long time for training and adaptation of the users. The myoelectric prostheses use signals from residual stump muscles to restore the function of the lost limbs seamlessly. However, the use of the sEMG signals in robotic as a user control signal is very complicated due to the presence of noise, and the need for heavy computational power. In this article, we developed motion intention classifiers for transradial (TR) amputees based on EMG data by implementing various machine learning and deep learning models. We benchmarked the performance of these classifiers based on overall generalization across various classes and we presented a systematic study on the impact of time domain features and pre-processing parameters on the performance of the classification models. Our results showed that Ensemble learning and deep learning algorithms outperformed other classical machine learning algorithms. Investigating the trend of varying sliding window on feature-based and non-feature-based classification model revealed interesting correlation with the level of amputation. The study also covered the analysis of performance of classifiers on amputation conditions since the history of amputation and conditions are different to each amputee. These results are vital for understanding the development of machine learning-based classifiers for assistive robotic applications.

Towards Robust, Adaptive and Reliable Upper-limb Motion Estimation Using Machine Learning and Deep Learning--A Survey in Myoelectric Control

To develop multi-functional human-machine interfaces that can help disabled people reconstruct lost functions of upper-limbs, machine learning (ML) and deep learning (DL) techniques have been widely implemented to decode human movement intentions from surface electromyography (sEMG) signals. However, due to the high complexity of upper-limb movements and the inherent non-stable characteristics of sEMG, the usability of ML/DL based control schemes is still greatly limited in practical scenarios. To this end, tremendous efforts have been made to improve model robustness, adaptation, and reliability. In this article, we provide a systematic review on recent achievements, mainly from three categories: multi-modal sensing fusion to gain additional information of the user, transfer learning (TL) methods to eliminate domain shift impacts on estimation models, and post-processing approaches to obtain more reliable outcomes. Special attention is given to fusion strategies, deep TL frameworks, and confidence estimation. \textcolor{red}{Research challenges and emerging opportunities, with respect to hardware development, public resources, and decoding strategies, are also analysed to provide perspectives for future developments.

DeepCov19Net: Automated COVID-19 Disease Detection with a Robust and Effective Technique Deep Learning Approach

The new type of coronavirus disease, which has spread from Wuhan, China since the beginning of 2020 called COVID-19, has caused many deaths and cases in most countries and has reached a global pandemic scale. In addition to test kits, imaging techniques with X-rays used in lung patients have been frequently used in the detection of COVID-19 cases. In the proposed method, a novel approach based on a deep learning model named DeepCovNet was utilized to classify chest X-ray images containing COVID-19, normal (healthy), and pneumonia classes. The convolutional-autoencoder model, which had convolutional layers in encoder and decoder blocks, was trained by using the processed chest X-ray images from scratch for deep feature extraction. The distinctive features were selected with a novel and robust algorithm named SDAR from the deep feature set. In the classification stage, an SVM classifier with various kernel functions was used to evaluate the classification performance of the proposed method. Also, hyperparameters of the SVM classifier were optimized with the Bayesian algorithm for increasing classification accuracy. Specificity, sensitivity, precision, and F-score, were also used as performance metrics in addition to accuracy which was used as the main criterion. The proposed method with an accuracy of 99.75 outperformed the other approaches based on deep learning.

Human motor function estimation based on EMG signal fractal dimension standard deviation

Wearable robots must adjust the assist mode/intensity according to human motion during the motion assistance process. By decoding the surface electromyography (sEMG) signal, the standard deviation of the fractal dimension is used as a characteristic index of muscle contraction-relaxation ability, and explore the feasibility of using the standard deviation of the fractal dimension to estimate the human motor function and thus provide a basis for decision-making for the flexible control of wearable robots. First, the sEMG signals of several subjects with different motor functions were collected and their time-domain and frequency-domain features were extracted. The experimental results for one hour of walking showed that the time-domain and frequency-domain feature values increased with muscle fatigue. The trend has little to do with the inherent motor function of the human body; Second, due to the strong nonlinearity, time-varying, and strong complexity of the sEMG signal, the fractal dimension nonlinear method is used to characterize the complexity of the EMG signal that is closely related to muscle function. Besides, theoretical and experimental studies have been conducted to clarify the feasibility of the complexity of fractal dimension representation and to provide theoretical support for the further use of the standard deviation of fractal dimension to estimate human motor function. The experimental results of continuous walking for one hour show that, macroscopically, the fractal dimension of each muscle of the individual subject does not change significantly with walking time, which shows that the fractal dimension has nothing to do with exercise time and muscle fatigue; On the microscopic level, the value of the fractal dimension changes when the subject’s muscles contract and relax. Subjects with strong motor function have smaller fractal dimensions when their muscles contract than subjects with weaker motor function, and the opposite happens when their muscles relax, and it can be seen that there is a positive correlation between the difference in the fractal dimension during muscle contraction and relaxation and the muscle contraction-relaxation ability and the human body’s inherent motor function. The test results verify the feasibility of using the standard deviation of fractal dimension to estimate the intrinsic motor function of the human body.

A Novel Medical Diagnosis model for COVID-19 infection detection based on Deep Features and Bayesian Optimization

A pneumonia of unknown causes, which was detected in Wuhan, China, and spread rapidly throughout the world, was declared as Coronavirus disease 2019 (COVID-19). Thousands of people have lost their lives to this disease. Its negative effects on public health are ongoing. In this study, an intelligence computer-aided model that can automatically detect positive COVID-19 cases is proposed to support daily clinical applications. The proposed model is based on the convolution neural network (CNN) architecture and can automatically reveal discriminative features on chest X-ray images through its convolution with rich filter families, abstraction, and weight-sharing characteristics. Contrary to the generally used transfer learning approach, the proposed deep CNN model was trained from scratch. Instead of the pre-trained CNNs, a novel serial network consisting of five convolution layers was designed. This CNN model was utilized as a deep feature extractor. The extracted deep discriminative features were used to feed the machine learning algorithms, which were k-nearest neighbor, support vector machine (SVM), and decision tree. The hyperparameters of the machine learning models were optimized using the Bayesian optimization algorithm. The experiments were conducted on a public COVID-19 radiology database. The database was divided into two parts as training and test sets with 70% and 30% rates, respectively. As a result, the most efficient results were ensured by the SVM classifier with an accuracy of 98.97%, a sensitivity of 89.39%, a specificity of 99.75%, and an F-score of 96.72%. Consequently, a cheap, fast, and reliable intelligence tool has been provided for COVID-19 infection detection. The developed model can be used to assist field specialists, physicians, and radiologists in the decision-making process. Thanks to the proposed tool, the misdiagnosis rates can be reduced, and the proposed model can be used as a retrospective evaluation tool to validate positive COVID-19 infection cases.