Recent trends and challenges of surface electromyography in prosthetic applications

Hand kinematics, high-density sEMG comprising forearm and far-field potentials for motion intent recognition

Surface electromyography (sEMG) signals reflect spinal motor neuron activities, and can be used as intuitive inputs for human-machine interaction (HMI) via movement intent recognition. The motor neuron potentials of far-field (wrist) and near-field (forearm) decomposed from high-density (HD) sEMG prospectively provide robust neural drives for HMI, which is a challenging research hotspot. However, there are no publicly available databases that include HD sEMG signals of forearm-wrist (FW) muscles, and hand kinematics (KIN). This paper presents the HD-FW KIN dataset that comprises HD 448-channel sEMG arrays distributed on forearm and wrist with simultaneously recording of finger joint angles and finger flexion forces. This dataset contains muscle activities of 21 subjects while performing 20 hand gestures, and 9 individual or combined finger flexion under two force levels. The usabilities of HD sEMG for hand gesture recognition, finger angle and force prediction were validated. The proposed database allows a comprehensive extraction of the neural drive from forearm and wrist, providing neural interfaces for the development of advanced prosthetic hand and wrist-worn consumer electronics.

Multimodal optimal matching and augmentation method for small sample gesture recognition

In human-computer interaction, gesture recognition based on physiological signals offers advantages such as a more natural and fast interaction mode and less constrained by the environment than visual-based. Surface electromyography-based gesture recognition has significantly progressed. However, since individuals have physical differences, researchers must collect data multiple times from each user to train the deep learning model. This data acquisition process can be particularly burdensome for non-healthy users. Researchers are currently exploring transfer learning and data augmentation techniques to enhance the accuracy of small-sample gesture recognition models. However, challenges persist, such as negative transfer and limited diversity in training samples, leading to suboptimal recognition performance. Therefore, We introduce motion information into sEMG-based recognition and propose a multimodal optimal matching and augmentation method for small sample gesture recognition, achieving efficient gesture recognition with only one acquisition per gesture. Firstly, this method utilizes the optimal matching signal selection module to select the most similar signals from the existing data to the new user as the training set, reducing inter-domain differences. Secondly, the similarity calculation augmentation module enhances the diversity of the training set. Finally, the Modal-type embedding enhances the information interaction between each mode signal. We evaluated the effectiveness on Self-collected Stroke Patient, the Ninapro DB1 dataset and the Ninapro DB5 dataset and achieved accuracies of 93.69%, 91.65% and 98.56%, respectively. These results demonstrate that the method achieved performance comparable to traditional recognition models while significantly reducing the collected data.

Improving movement decoding performance under joint constraints based on a neural-driven musculoskeletal model

Electromyography-driven musculoskeletal model (E-DMM) connects the user's control commands with the joint positions from a physiological perspective. However, features extracted directly from the surface EMG signals may be affected by signal crosstalk and amplitude cancellation. This limitation can be addressed with the decomposition algorithms for high-density (HD) EMG signals, which demonstrate the capability of extracting neural drives for the human-machine interface. On this basis, we proposed a neural-driven musculoskeletal model (N-DMM) with improved movement decoding performance for estimating wrist and metacarpophalangeal (MCP) joint positions under joint constraints. Eight limb-intact subjects participated in the experiment of mirrored bilateral training. The wrist and MCP joints of the subjects on one side were constrained, and the HD EMG signals from the same side were recorded. Moreover, the unconstrained side mirrored the joint movements of the phantom limb, while the joint angles were measured simultaneously. The obtained EMG signals were processed with the fast independent component analysis algorithm to extract motor unit discharges, enabling the estimation of neural drives. Then the neural drives were taken as inputs for the N-DMM to estimate joint movements. For comparison, an E-DMM was also employed for joint angle prediction. The results indicated that our N-DMM demonstrated superior performance compared to the E-DMM, potentially allowing for more accurate and robust decoding of continuous movements under joint constraints. Further improvement of the proposed model could offer a promising approach for practical applications in amputees.

Machine Learning- and Deep Learning-Based Myoelectric Control System for Upper Limb Rehabilitation Utilizing EEG and EMG Signals: A Systematic Review

Upper limb disabilities, often caused by conditions such as stroke or neurological disorders, severely limit an individual's ability to perform essential daily tasks, leading to a significant reduction in quality of life. The development of effective rehabilitation technologies is crucial to restoring motor function and improving patient outcomes. This systematic review examines the application of machine learning and deep learning techniques in myoelectric-controlled systems for upper limb rehabilitation, focusing on the use of electroencephalography and electromyography signals. By integrating non-invasive signal acquisition methods with advanced computational models, the review highlights how these technologies can enhance the accuracy and efficiency of rehabilitation devices. A comprehensive search of literature published between January 2015 and July 2024 led to the selection of fourteen studies that met the inclusion criteria. These studies showcase various approaches in decoding motor intentions and controlling assistive devices, with models such as Long Short-Term Memory Networks, Support Vector Machines, and Convolutional Neural Networks showing notable improvements in control precision. However, challenges remain in terms of model robustness, computational complexity, and real-time applicability. This systematic review aims to provide researchers with a deeper understanding of the current advancements and challenges in this field, guiding future research efforts to overcome these barriers and facilitate the transition of these technologies from experimental settings to practical, real-world applications.

The neurophysiology of sensorimotor prosthetic control

Movement is a central behavior of daily living; thus lost or compromised movement due to disease, injury, or amputation causes enormous loss of productivity and quality of life. While prosthetics have evolved enormously over the years, restoring natural sensorimotor (SM) control via a prosthesis is a difficult problem which neuroengineering has yet to solve. With a focus on upper limb prosthetics, this perspective article discusses the neurophysiology of motor control under healthy conditions and after amputation, the development of upper limb prostheses from early generations to current state-of-the art sensorimotor neuroprostheses, and how postinjury changes could complicate prosthetic control. Current challenges and future development of smart sensorimotor neuroprostheses are also discussed.

Developing a device for simultaneously investigating pivoting neuromuscular control and muscle properties toward a multi-axis rehabilitation

Understanding the pivoting neuromuscular control of the lower limb and its associated muscle properties is critical for developing diagnostic and rehabilitation tools. However, to the best of our knowledge, a device that can evaluate these factors simultaneously remains lacking. To address this gap, a device that can investigate pivoting neuromuscular control and associated muscle properties was developed in this study. The proposed device consisted of a pivoting mechanism and height-adjustable chair with a brace interface. The device can control a footplate at various speeds to facilitate pivoting stretching and quantify neuromuscular control. Time-synchronized ultrasonographic images can be acquired simultaneously to quantify muscle properties during both active and passive pivoting movements. The muscle displacement, fascicle length/displacement, pennation angle, pivoting stiffness, and pivoting instability were investigated using the proposed device. Further, the feasibility of the device was demonstrated through a cross-sectional study with healthy subjects. The proposed device successfully quantified changes in muscle displacement during passive and active pivoting movements, pivoting stiffness during passive movements, and neuromuscular control during active movements. Therefore, the proposed device is expected to be used as a research and therapeutic tool for improving pivoting neuromuscular control and muscle functions and investigating the underlying mechanisms associated between muscle properties and joint movement in the transverse plane.

Analysis of induced dynamic biceps EMG signal complexity using Markov transition networks

Purpose

Surface electromyography (sEMG) is a non-invasive technique to characterize muscle electrical activity. The analysis of sEMG signals under muscle fatigue play a crucial part in the branch of neurorehabilitation, sports medicine, biomechanics, and monitoring neuromuscular pathologies. In this work, a method to transform sEMG signals to complex networks under muscle fatigue conditions using Markov transition field (MTF) is proposed. The importance of normalization to a constant Maximum voluntary contraction (MVC) is also considered.

Methods

For this, dynamic signals are recorded using two different experimental protocols one under constant load and another referenced to 50% MVC from Biceps brachii of 50 and 45 healthy subjects respectively. MTF is generated and network graph is constructed from preprocesses signals. Features such as average self-transition probability, average clustering coefficient and modularity are extracted.

Results

All the extracted features showed statistical significance for the recorded signals. It is found that during the transition from non-fatigue to fatigue, average clustering coefficient decreases while average self-transition probability and modularity increases.

Conclusion

The results indicate higher degree of signal complexity during non-fatigue condition. Thus, the MTF approach may be used to indicate the complexity of sEMG signals. Although both datasets showed same trend in results, sEMG signals under 50% MVC exhibited higher separability for the features. The inter individual variations of the MTF features is found to be more for the signals recorded using constant load. The proposed study can be adopted to study the complex nature of muscles under various neuromuscular conditions.

A Prototype System for High Frame Rate Ultrasound Imaging based Prosthetic Arm Control

The creation of unique control methods for a hand prosthesis is still a problem that has to be addressed. The best choice of a human-machine interface (HMI) that should be used to enable natural control is still a challenge. Surface electromyography (sEMG), the most popular option, has a variety of difficult-to-fix issues (electrode displacement, sweat, fatigue). The ultrasound imaging-based methodology offers a means of recognising complex muscle activity and configuration with a greater SNR and less hardware requirements as compared to sEMG. In this study, a prototype system for high frame rate ultrasound imaging for prosthetic arm control is proposed. Using the proposed framework, a virtual robotic hand simulation is developed that can mimic a human hand as illustrated in the link: https://youtu.be/LBcwQ0xzQK0. The proposed classification model simulating four hand gestures has a classification accuracy of more than 90%.Clinical relevance-The proposed system enables an ultrasound imaging based human machine interface that can be a research and development platform for novel control strategies of a hand prosthesis.