Detection of Multiple Innervation Zones from Multi-Channel Surface EMG Recordings with Low Signal-to-Noise Ratio Using Graph-Cut Segmentation

A method for the estimation of a motor unit innervation zone center position evaluated with a computational sEMG model

Motion predictions for limbs can be performed using commonly called Hill-based muscle models. For this type of models, a surface electromyogram (sEMG) of the muscle serves as an input signal for the activation of the muscle model. However, the Hill model needs additional information about the mechanical system state of the muscle (current length, velocity, etc.) for a reliable prediction of the muscle force generation and, hence, the prediction of the joint motion. One feature that contains potential information about the state of the muscle is the position of the center of the innervation zone. This feature can be further extracted from the sEMG. To find the center, a wavelet-based algorithm is proposed that localizes motor unit potentials in the individual channels of a single-column sEMG array and then identifies innervation point candidates. In the final step, these innervation point candidates are clustered in a density-based manner. The center of the largest cluster is the estimated center of the innervation zone. The algorithm has been tested in a simulation. For this purpose, an sEMG simulator was developed and implemented that can compute large motor units (1,000's of muscle fibers) quickly (within seconds on a standard PC).

Muscle innervation zone estimation from monopolar high-density M-waves using principal component analysis and radon transform

This study examined methods for estimating the innervation zone (IZ) of a muscle using recorded monopolar high density M waves. Two IZ estimation methods based on principal component analysis (PCA) and Radon transform (RT) were examined. Experimental M waves, acquired from the biceps brachii muscles of nine healthy subjects were used as testing data sets. The performance of the two methods was evaluated by comparing their IZ estimations with manual IZ detection by experienced human operators. Compared with manual detection, the agreement rate of the estimated IZs was 83% and 63% for PCA and RT based methods, respectively, both using monopolar high density M waves. In contrast, the agreement rate was 56% for cross correlation analysis using bipolar high density M waves. The mean difference in estimated IZ location between manual detection and the tested method was 0.12 ± 0.28 inter-electrode-distance (IED) for PCA, 0.33 ± 0.41 IED for RT and 0.39 ± 0.74 IED for cross correlation-based methods. The results indicate that the PCA based method was able to automatically detect muscle IZs from monopolar M waves. Thus, PCA provides an alternative approach to estimate IZ location of voluntary or electrically-evoked muscle contractions, and may have particular value for IZ detection in patients with impaired voluntary muscle activation.

Kernel Density Estimation of Electromyographic Signals and Ensemble Learning for Highly Accurate Classification of a Large Set of Hand/Wrist Motions

The performance of myoelectric control highly depends on the features extracted from surface electromyographic (sEMG) signals. We propose three new sEMG features based on the kernel density estimation. The trimmed mean of density (TMD), the entropy of density, and the trimmed mean absolute value of derivative density were computed for each sEMG channel. These features were tested for the classification of single tasks as well as of two tasks concurrently performed. For single tasks, correlation-based feature selection was used, and the features were then classified using linear discriminant analysis (LDA), non-linear support vector machines, and multi-layer perceptron. The eXtreme gradient boosting (XGBoost) classifier was used for the classification of two movements simultaneously performed. The second and third versions of the Ninapro dataset (conventional control) and Ameri's movement dataset (simultaneous control) were used to test the proposed features. For the Ninapro dataset, the overall accuracy of LDA using the TMD feature was 98.99 ± 1.36% and 92.25 ± 9.48% for able-bodied and amputee subjects, respectively. Using ensemble learning of the three classifiers, the average macro and micro-F-score, macro recall, and precision on the validation sets were 98.23 ± 2.02, 98.32 ± 1.93, 98.32 ± 1.93, and 98.88 ± 1.31%, respectively, for the intact subjects. The movement misclassification percentage was 1.75 ± 1.73 and 3.44 ± 2.23 for the intact subjects and amputees. The proposed features were significantly correlated with the movement classes [Generalized Linear Model (GLM); P-value < 0.05]. An accurate online implementation of the proposed algorithm was also presented. For the simultaneous control, the overall accuracy was 99.71 ± 0.08 and 97.85 ± 0.10 for the XGBoost and LDA classifiers, respectively. The proposed features are thus promising for conventional and simultaneous myoelectric control.

Distribution of innervation zone and muscle fiber conduction velocity in the biceps brachii muscle

The spatial distributions of muscle innervation zone (IZ) and muscle fiber conduction velocity (CV) were examined in nine healthy young male participants. High-density surface electromyography (EMG) was collected from the biceps brachii muscle when subjects performed isometric elbow flexions at 20% to 80% of the maximal voluntary contraction (MVC). A total of 9498 samples of IZs were identified and CVs were calculated using the Radon transform. The center and width of IZ sample distribution were compared within four different force levels and six medial to lateral electrode column positions using repeated measures ANOVA and multiple comparison tests. Significant shifts of IZ center were observed in the medial columns (Columns 5, 6, and 7) compared with the lateral columns (Columns 3 and 4) (p < 0.05). Similarly, significant differences in the IZ width were found in Column 7 and 8 compared to Column 3 (p < 0.05). In contrast, muscle CV was unaffected by column position. Instead, muscle CV was faster at 40% and 80% MVC compared to 20% MVC (p < 0.05). The findings of this study add further insights into the physiological properties of the biceps brachii muscle.

Neurophysiological Factors Affecting Muscle Innervation Zone Estimation Using Surface EMG: A Simulation Study

Surface electromyography (EMG) recorded by a linear or 2-dimensional electrode array can be used to estimate the location of muscle innervation zones (IZ). There are various neurophysiological factors that may influence surface EMG and thus potentially compromise muscle IZ estimation. The objective of this study was to evaluate how surface-EMG-based IZ estimation might be affected by different factors, including varying degrees of motor unit (MU) synchronization in the case of single or double IZs. The study was performed by implementing a model simulating surface EMG activity. Three different MU synchronization conditions were simulated, namely no synchronization, medium level synchronization, and complete synchronization analog to M wave. Surface EMG signals recorded by a 2-dimensional electrode array were simulated from a muscle with single and double IZs, respectively. For each situation, the IZ was estimated from surface EMG and compared with the one used in the model for performance evaluation. For the muscle with only one IZ, the estimated IZ location from surface EMG was consistent with the one used in the model for all the three MU synchronization conditions. For the muscle with double IZs, at least one IZ was appropriately estimated from interference surface EMG when there was no MU synchronization. However, the estimated IZ was different from either of the two IZ locations used in the model for the other two MU synchronization conditions. For muscles with a single IZ, MU synchronization has little effect on IZ estimation from electrode array surface EMG. However, caution is required for multiple IZ muscles since MU synchronization might lead to false IZ estimation.

High-density surface electromyography signals during isometric contractions of elbow muscles of healthy humans

This paper presents a dataset of high-density surface EMG signals (HD-sEMG) designed to study patterns of sEMG spatial distribution over upper limb muscles during voluntary isometric contractions. Twelve healthy subjects performed four different isometric tasks at different effort levels associated with movements of the forearm. Three 2-D electrode arrays were used for recording the myoelectric activity from five upper limb muscles: biceps brachii, triceps brachii, anconeus, brachioradialis, and pronator teres. Technical validation comprised a signals quality assessment from outlier detection algorithms based on supervised and non-supervised classification methods. About 6% of the total number of signals were identified as "bad" channels demonstrating the high quality of the recordings. In addition, spatial and intensity features of HD-sEMG maps for identification of effort type and level, have been formulated in the framework of this database, demonstrating better performance than the traditional time-domain features. The presented database can be used for pattern recognition and MUAP identification among other uses.

Multi-Channel Surface EMG Spatio-Temporal Image Enhancement Using Multi-Scale Hessian-Based Filters

Surface electromyography (sEMG) signals acquired with linear electrode array are useful in analyzing muscle anatomy and physiology. Most algorithms for signal processing, detection, and estimation require adequate quality of the input signals, however, multi-channel sEMG signals are commonly contaminated due to several noise sources. The sEMG signal needs to be enhanced prior to the digital signal and image processing to achieve the best results. This study is using spatio-temporal images to represent surface EMG signals. The motor unit action potential (MUAP) in these images looks like a linear structure, making certain angles with the x-axis, depending on the conduction velocity of the MU. A multi-scale Hessian-based filter is used to enhance the linear structure, i.e., the MUAP region, and to suppress the background noise. The proposed framework is compared with some of the existing algorithms using synthetic, simulated, and experimental sEMG signals. Results show improved detection accuracy of the motor unit action potential after the proposed enhancement as a preprocessing step.

Upper extremity surface electromyography signal changes after laparoscopic training

Introduction

Objective measures of laparoscopic skill in training are lacking.

Aim

To evaluate the changes in the surface electromyography (sEMG) signal during laparoscopic training, and to link them to intracorporeal knot tying.

Material and methods

Ten right-handed medical students (6 female), aged 25 ±0.98, without training in laparoscopy, were enrolled in the study. With no additional training, they tied intracorporeal single knots for 15 min. Then underwent laparoscopic training and redid the knot tying exercise. During both events, sEMG was recorded from 8 measurement points on the upper extremities and neck bilaterally. We analyzed changes in sEMG resulting from training and tried to find sEMG predictive parameters for higher technical competence defined by the number of knots tied after the training.

Results

The average number of knots increased after the training. Significant decreases in activity after the training were visible for the non-dominant hand deltoid and trapezius muscles. Dominant and non-dominant hands had different activation patterns. Differences largely disappeared after the training. All muscles, except for the dominant forearm and non-dominant thenar, produced a negative correlation between their activities and the number of tied knots. The strongest anticorrelation occurred for the non-dominant deltoid (r = –0.863, p < 0.05). Relatively strong relationships were identified in the case of the non-dominant trapezius and forearm muscles (r = –0.587, r = –0.504).

Conclusions

At least for some muscle groups there is a change in activation patterns after laparoscopic training. Proximal muscle groups tend to become more relaxed and the distal ones become more active. Changes in the non-dominant hand are more pronounced than in the dominant hand.

Motor-commands decoding using peripheral nerve signals: a review

During the last few decades, substantial scientific and technological efforts have been focused on the development of neuroprostheses. The major emphasis has been on techniques for connecting the human nervous system with a robotic prosthesis via natural-feeling interfaces. The peripheral nerves provide access to highly processed and segregated neural command signals from the brain that can in principle be used to determine user intent and control muscles. If these signals could be used, they might allow near-natural and intuitive control of prosthetic limbs with multiple degrees of freedom. This review summarizes the history of neuroprosthetic interfaces and their ability to record from and stimulate peripheral nerves. We also discuss the types of interfaces available and their applications, the kinds of peripheral nerve signals that are used, and the algorithms used to decode them. Finally, we explore the prospects for future development in this area.

A Novel Spatial Feature for the Identification of Motor Tasks Using High-Density Electromyography

Estimation of neuromuscular intention using electromyography (EMG) and pattern recognition is still an open problem. One of the reasons is that the pattern-recognition approach is greatly influenced by temporal changes in electromyograms caused by the variations in the conductivity of the skin and/or electrodes, or physiological changes such as muscle fatigue. This paper proposes novel features for task identification extracted from the high-density electromyographic signal (HD-EMG) by applying the mean shift channel selection algorithm evaluated using a simple and fast classifier-linear discriminant analysis. HD-EMG was recorded from eight subjects during four upper-limb isometric motor tasks (flexion/extension, supination/pronation of the forearm) at three different levels of effort. Task and effort level identification showed very high classification rates in all cases. This new feature performed remarkably well particularly in the identification at very low effort levels. This could be a step towards the natural control in everyday applications where a subject could use low levels of effort to achieve motor tasks. Furthermore, it ensures reliable identification even in the presence of myoelectric fatigue and showed robustness to temporal changes in EMG, which could make it suitable in long-term applications.