Myoelectric Control Algorithms Research Articles

EMG Dataset for Gesture Recognition with Arm Translation

Myoelectric control has emerged as a promising approach for a wide range of applications, including controlling limb prosthetics, teleoperating robots and enabling immersive interactions in the Metaverse. However, the accuracy and robustness of myoelectric control systems are often affected by various factors, including muscle fatigue, perspiration, drifts in electrode positions and changes in arm position. The latter has received less attention despite its significant impact on signal quality and decoding accuracy. To address this gap, we present a novel dataset of surface electromyographic (EMG) signals captured from multiple arm positions. This dataset, comprising EMG and hand kinematics data from 8 participants performing 6 different hand gestures, provides a comprehensive resource for investigating position-invariant myoelectric control decoding algorithms. We envision this dataset to serve as a valuable resource for both training and benchmark arm position-invariant myoelectric control algorithms. Additionally, to expand the publicly available data capturing the variability of EMG signals across diverse arm positions, we propose a novel data acquisition protocol that can be utilized for future data collection.

Validity and Impact of Methods for Collecting Training Data for Myoelectric Prosthetic Control Algorithms.

Intuitive regression control of prostheses relies on training algorithms to correlate biological recordings to motor intent. The quality of the training dataset is critical to run-time regression performance, but accurately labeling intended hand kinematics after hand amputation is challenging. In this study, we quantified the accuracy and precision of labeling hand kinematics using two common training paradigms: 1) mimic training, where participants mimic predetermined motions of a prosthesis, and 2) mirror training, where participants mirror their contralateral intact hand during synchronized bilateral movements. We first explored this question in healthy non-amputee individuals where the ground-truth kinematics could be readily determined using motion capture. Kinematic data showed that mimic training fails to account for biomechanical coupling and temporal changes in hand posture. Additionally, mirror training exhibited significantly higher accuracy and precision in labeling hand kinematics. These findings suggest that the mirror training approach generates a more faithful, albeit more complex, dataset. Accordingly, mirror training resulted in significantly better offline regression performance when using a large amount of training data and a non-linear neural network. Next, we explored these different training paradigms online, with a cohort of unilateral transradial amputees actively controlling a prosthesis in real-time to complete a functional task. Overall, we found that mirror training resulted in significantly faster task completion speeds and similar subjective workload. These results demonstrate that mirror training can potentially provide more dexterous control through the utilization of task-specific, user-selected training data. Consequently, these findings serve as a valuable guide for the next generation of myoelectric and neuroprostheses leveraging machine learning to provide more dexterous and intuitive control.

Efficient correction of armband rotation for myoelectric-based gesture control interface

Objective. The appearance of commercial myoelectric armbands has greatly increased the portability and convenience of myoelectric controlled interfaces (MCIs). However, one limitation of the current state-of-the-art myoelectric control algorithms is that they have poor robustness against armband displacements, especially rotation, leading to great algorithmic performance degradation. The traditional remedy, retraining the interface, requires the data collection of all gestures and is impractical in many applications. The recently proposed position verification (PV) framework focused on quickly identifying and correcting the electrode positions after the displacement, showing the potential to restore the performance of MCI in a faster way. However, its online effectiveness is still yet to be validated. Approach. This work proposed a novel algorithm of identifying the rotation direction to improve the efficiency of the PV framework and demonstrated the real-time capability of the PV framework using a commercially available armband. Main results. The results showed that with PV, a 1.5-cm rotation could be corrected with an average of 3.1 ± 1.5 interactive adjustments, equivalent to around 15.5 ± 7.5 s, which was greatly reduced compared to retraining. There was no significant difference in the real-time control performance between before the armband displacement and after the PV correction. Significance. To the best of our knowledge, this study was the first maintaining pattern recognition-based myoelectric control performance in the presence of electrode shifts without recollecting the entire training data. It suggested the feasibility of the PV framework used in the myoelectric armband and MCI for practical applications.

A Novel Framework Based on Position Verification for Robust Myoelectric Control Against Sensor Shift

This study presented a novel framework to improve the robustness of pattern recognition-based myoelectric control algorithms against sensor shift, which was one of the obstacles for its practical applications outside controlled laboratory conditions. Different from the previous proposed methods, which mostly focused on improving the classification performance in the shift condition, this framework provided the functionality of verifying if the sensor position was shifted. If so, the user was enabled to adjust the sensor position before it affecting the following control. We demonstrated that one verification attempt could be completed in a short time (< 2 s) with a high accuracy (equal error rate, or EER< 5%). The control performance was theoretically proved to be improved after position verification. The simulated results of the experimental data from six sensors around the arm showed that with improved discrete Fourier transform (iDFT) feature, in most scenarios, smaller than five verification attempts (< 10 s) were needed to correct the sensor position from 1-cm shift perpendicular to muscle fibers. The performance after correction was comparable to that of a traditional method including additional training data from the expected shift positions. The proposed framework dealt with the sensor shift problem from the perspective of efficient position correction, potentially expanding the application of myoelectric control, especially with armband, in industry.

Effect of User Practice on Prosthetic Finger Control With an Intuitive Myoelectric Decoder.

Machine learning-based myoelectric control is regarded as an intuitive paradigm, because of the mapping it creates between muscle co-activation patterns and prosthesis movements that aims to simulate the physiological pathways found in the human arm. Despite that, there has been evidence that closed-loop interaction with a classification-based interface results in user adaptation, which leads to performance improvement with experience. Recently, there has been a focus shift toward continuous prosthesis control, yet little is known about whether and how user adaptation affects myoelectric control performance in dexterous, intuitive tasks. We investigate the effect of short-term adaptation with independent finger position control by conducting real-time experiments with 10 able-bodied and two transradial amputee subjects. We demonstrate that despite using an intuitive decoder, experience leads to significant improvements in performance. We argue that this is due to the lack of an utterly natural control scheme, which is mainly caused by differences in the anatomy of human and artificial hands, movement intent decoding inaccuracies, and lack of proprioception. Finally, we extend previous work in classification-based and wrist continuous control by verifying that offline analyses cannot reliably predict real-time performance, thereby reiterating the importance of validating myoelectric control algorithms with real-time experiments.

Predicting wrist kinematics from motor unit discharge timings for the control of active prostheses

BackgroundCurrent myoelectric control algorithms for active prostheses map time- and frequency-domain features of the interference EMG signal into prosthesis commands. With this approach, only a fraction of the available information content of the EMG is used and the resulting control fails to satisfy the majority of users. In this study, we predict joint angles of the three degrees of freedom of the wrist from motor unit discharge timings identified by decomposition of high-density surface EMG.MethodsWe recorded wrist kinematics and high-density surface EMG signals from six able-bodied individuals and one patient with limb deficiency while they performed movements of three degrees of freedom of the wrist at three different speeds. We compared the performance of linear regression to predict the observed individual wrist joint angles from, either traditional time domain features of the interference EMG or from motor unit discharge timings (which we termed neural features) obtained by EMG decomposition. In addition, we propose and test a simple model-based dimensionality reduction, based on the physiological notion that the discharge timings of motor units are partly correlated.ResultsThe regression approach using neural features outperformed regression on classic global EMG features (average R2 for neural features 0.77 and 0.64, for able-bodied subjects and patients, respectively; for time-domain features 0.70 and 0.52).ConclusionsThese results indicate that the use of neural information extracted from EMG decomposition can advance man-machine interfacing for prosthesis control.

Electrode Density Affects the Robustness of Myoelectric Pattern Recognition System With and Without Electrode Shift.

With the availability of high-density (HD) electrodes technology, the electrodes used in myoelectric control can have much higher density than the current practice. In this study, we investigated the effects of electrode density on pattern recognition (PR) based myoelectric control. Four density levels were analyzed in two directions: parallel and perpendicular to muscle fibers. Their influence on PR-based myoelectric control algorithms was investigated under three conditions between training and testing datasets: no electrode shift, 10-mm shift parallel to muscle fibers and 10-mm shift perpendicular to muscle fibers. The effect of electrode density varied among the different shift conditions: First, when there was no shift, increasing electrode density significantly improved the classification performance; second, when the shift was in the perpendicular direction, increasing electrode density resulted in deterioration in the classification performance; third, when the shift was in the parallel direction, the effect of the electrode density was more complicated-increasing the density in the parallel direction reduced the performance, while increasing density in the perpendicular direction would initially enhance the performance, but then reduce performance. To our best knowledge, this was the first study focusing on the role of electrode density in myoelectric control with the presence of electrode shift. Its outcome would benefit the design of electrode placement for future myoelectric prostheses with HD electrodes.

Towards Zero Retraining for Myoelectric Control Based on Common Model Component Analysis.

In spite of several decades of intense research and development, the existing algorithms of myoelectric pattern recognition (MPR) are yet to satisfy the criteria that a practical upper extremity prostheses should fulfill. This study focuses on the criterion of the short, or even zero subject training. Due to the inherent nonstationarity in surface electromyography (sEMG) signals, current myoelectric control algorithms usually need to be retrained daily during a multiple days' usage. This study was conducted based on the hypothesis that there exist some invariant characteristics in the sEMG signals when a subject performs the same motion in different days. Therefore, given a set of classifiers (models) trained on several days, it is possible to find common characteristics among them. To this end, we proposed to use common model component analysis (CMCA) framework, in which an optimized projection was found to minimize the dissimilarity among multiple models of linear discriminant analysis (LDA) trained using data from different days. Five intact-limbed subjects and two transradial amputee subjects participated in an experiment including six sessions of sEMG data recording, which were performed in six different days, to simulate the application of MPR over multiple days. The results demonstrate that CMCA has a significant better generalization ability with unseen data (not included in the training data), leading to classification accuracy improvement and increase of completion rate in a motion test simulation, when comparing with the baseline reference method. The results indicate that CMCA holds a great potential in the effort of developing zero retraining of MPR.

Is Accurate Mapping of EMG Signals on Kinematics Needed for Precise Online Myoelectric Control?

In this paper, we present a systematic analysis of the relationship between the accuracy of the mapping between EMG and hand kinematics and the control performance in goal-oriented tasks of three simultaneous and proportional myoelectric control algorithms: nonnegative matrix factorization (NMF), linear regression (LR), and artificial neural networks (ANN). The purpose was to investigate the impact of the precision of the kinematics estimation by a myoelectric controller for accurately complete goal-directed tasks. Nine naïve subjects performed a series of goal-directed myoelectric control tasks using the three algorithms, and their online performance was characterized by 6 indexes. The results showed that, although the three algorithms' mapping accuracies were significantly different, their online performance was similar. Moreover, for LR and ANN, the offline performance was not correlated to any of the online performance indexes, and only a weak correlation was found with three of them for NMF . We conclude that for reliable simultaneous and proportional myoelectric control, it is not necessary to achieve high accuracy in the mapping between EMG and kinematics. Rather, good online myoelectric control is achieved by the continuous interaction and adaptation of the user with the myoelectric controller through feedback (visual in the current study). Control signals generated by EMG with rather poor association with kinematic variables can still be fully exploited by the user for precise control. This conclusion explains the possibility of accurate simultaneous and proportional control over multiple degrees of freedom when using unsupervised algorithms, such as NMF.

Myoelectric Walking Mode Classification for Transtibial Amputees

Myoelectric control algorithms have the potential to detect an amputee's motion intent and allow the prosthetic to adapt to changes in walking mode. The development of a myoelectric walking mode classifier for transtibial amputees is outlined. Myoelectric signals from four muscles (tibialis anterior, medial gastrocnemius (MG), vastus lateralis, and biceps femoris) were recorded for five nonamputee subjects and five transtibial amputees over a variety of walking modes: level ground at three speeds, ramp ascent/descent, and stair ascent/descent. These signals were decomposed into relevant features (mean absolute value, variance, wavelength, number of slope sign changes, number of zero crossings) over three subwindows from the gait cycle and used to test the ability of classification algorithms for transtibial amputees using linear discriminant analysis (LDA) and support vector machine (SVM) classifiers. Detection of all seven walking modes had an accuracy of 97.9% for the amputee group and 94.7% for the nonamputee group. Misclassifications occurred most frequently between different walking speeds due to the similar nature of the gait pattern. Stair ascent/descent had the best classification accuracy with 99.8% for the amputee group and 100.0% for the nonamputee group. Stability of the developed classifier was explored using an electrode shift disturbance for each muscle. Shifting the electrode placement of the MG had the most pronounced effect on the classification accuracy for both samples. No increase in classification accuracy was observed when using SVM compared to LDA for the current dataset.