Search

This study aims to explore alterations in corticomuscular and cortical coupling during the rehabilitation of stroke patients. We initiated the analysis by employing variational modal decomposition (VMD) on electromyography (EMG) data, followed by the application of VDM-transfer entropy (VMD-TE) to quantify the coupling strength between electroencephalogram (EEG) and EMG signals. Subsequently, we constructed the VMD-TE connection matrix and analyzed the clustering coefficient and small-world attributes within the cortico-muscular functional network (CMFN). Finally, a random forest algorithm was employed to extract features from the VMD-TE connection matrix across different rehabilitation periods. Beta waves in EEG were emerged as the key information carrier between the cortex and muscle, and the CMFN of patients with the beta frequency band has small-world characteristics. During rehabilitation, we observed a decrease in coupling between the initially affected motor cortex and muscle, accompanied by an increase in coupling between the frontal region and muscle. Our findings suggest potential neuro-remodeling in stroke patients after rehabilitation, with CFMN serving as a valuable metric for assessing cortico-muscular coupling.

ObjectiveNowadays, increasingly studies are attempting to analyze strokes in advance. The identification of brain damage areas is essential for stroke rehabilitation.ApproachWe proposed Electroencephalogram (EEG) multi-modal frequency features to classify the regions of stroke injury. The EEG signals were obtained from stroke patients and healthy subjects, who were divided into right-sided brain injury group, left-sided brain injury group, bilateral brain injury group, and healthy controls. First, the wavelet packet transform was used to perform a time-frequency analysis of the EEG signal and extracted a set of features (denoted as WPT features). Then, to explore the nonlinear phase coupling information of the EEG signal, phase-locked values (PLV) and partial directed correlations (PDC) were extracted from the brain network, and the brain network produced a second set of features noted as functional connectivity (FC) features. Furthermore, we fused the extracted multiple features and used the resnet50 convolutional neural network to classify the fused multi-modal (WPT + FC) features.ResultsThe classification accuracy of our proposed methods was up to 99.75%.SignificanceThe proposed multi-modal frequency features can be used as a potential indicator to distinguish regions of brain injury in stroke patients, and are potentially useful for the optimization of decoding algorithms for brain-computer interfaces.

Changes in the functional connections between the cerebral cortex and muscles can evaluate motor function in stroke rehabilitation. To quantify changes in functional connections between the cerebral cortex and muscles, we combined corticomuscular coupling and graph theory to propose dynamic time warped (DTW) distances for electroencephalogram (EEG) and electromyography (EMG) signals as well as two new symmetry metrics. EEG and EMG data from 18 stroke patients and 16 healthy individuals, as well as Brunnstrom scores from stroke patients, were recorded in this paper. First, calculate DTW-EEG, DTW-EMG, BNDSI and CMCSI. Then, the random forest algorithm was used to calculate the feature importance of these biological indicators. Finally, based on the results of feature importance, different features were combined and validated for classification. The results showed that the feature importance was from high to low as CMCSI/BNDSI/DTW-EEG/DTW-EMG, while the feature combination with the highest accuracy was CMCSI+BNDSI+DTW-EEG. Compared to previous studies, combining the CMCSI+BNDSI+DTW-EEG features of EEG and EMG achieved better results in the prediction of motor function rehabilitation at different levels of stroke. Our work implies that the establishment of a symmetry index based on graph theory and cortical muscle coupling has great potential in predicting stroke recovery and promises to have an impact on clinical research applications.

Motor imagery (MI) electroencephalogram (EEG) signals have an important role in brain-computer interface (BCI) research. However, effectively decoding these signals remains a problem to be solved. Traditional EEG signal decoding algorithms rely on parameter design to extract features, whereas deep learning algorithms represented by convolution neural network (CNN) can automatically extract features, which is more suitable for BCI applications. However, when EEG data is taken as input in raw time series, traditional 1D-CNNs are unable to acquire both frequency domain and channel association information. To solve this problem, this study proposes a novel algorithm by inserting two modules into CNN. One is the Filter Band Combination (FBC) Module, which preserves as many frequency domain features as possible while maintaining the time domain characteristics of EEG. Another module is Multi-View structure that can extract features from the output of FBC module. To prevent over fitting, we used a cosine annealing algorithm with restart strategy to update the learning rate. The proposed algorithm was validated on the BCI competition dataset and the experiment dataset, using accuracy, standard deviation, and kappa coefficient. Compared with traditional decoding algorithms, our proposed algorithm achieved an improvement of the maximum average correct rate of 6.6% on the motion imagery 4-classes recognition mission and 11.3% on the 2-classes classification task.

IntroductionRestless legs syndrome (RLS) is a common sensorimotor disorder characterized by an irrepressible urge to move the legs and frequently accompanied by unpleasant sensations in the legs. The pathophysiological mechanisms underlying RLS remain unclear, and RLS is hypothesized to be associated with alterations in white matter tracts.MethodsDiffusion MRI is a unique noninvasive method widely used to study white matter tracts in the human brain. Thus, diffusion-weighted images were acquired from 18 idiopathic RLS patients and 31 age- and sex-matched healthy controls (HCs). Whole brain tract-based spatial statistics (TBSS) and atlas-based analyzes combining crossing fiber-based metrics and tensor-based metrics were performed to investigate the white matter patterns in individuals with RLS.ResultsTBSS analysis revealed significantly higher fractional anisotropy (FA) and partial volume fraction of primary (F1) fiber populations in multiple tracts associated with the sensorimotor network in patients with RLS than in HCs. In the atlas based analysis, the bilateral anterior thalamus radiation, bilateral corticospinal tract, bilateral inferior fronto-occipital fasciculus, left hippocampal cingulum, left inferior longitudinal fasciculus, and left uncinate fasciculus showed significantl increased F1, but only the left hippocampal cingulum showed significantly higher FA.DiscussionThe results demonstrated that F1 identified extensive alterations in white matter tracts compared with FA and confirmed the hypothesis that crossing fiber-based metrics are more sensitive than tensor-based metrics in detecting white matter abnormalities in RLS. The present findings provide evidence that the increased F1 metric observed in sensorimotor tracts may be a critical neural substrate of RLS, enhancing our understanding of the underlying pathological changes.

Sign language (SL) has strong structural features. Various gestures and the complex trajectories of hand movements bring challenges to sign language recognition (SLR). Based on the inherent correlation between gesture and trajectory of SL action, SLR is organically divided into gesture-based recognition and gesture-related movement trajectory recognition. One hundred and twenty commonly used Chinese SL words involving 9 gestures and 8 movement trajectories, are selected as research and test objects. The method based on the amplitude state of surface electromyography (sEMG) signal and acceleration signal is used for vocabulary segmentation. The multi-sensor decision fusion method of coupled hidden Markov model is used to complete the recognition of SL vocabulary, and the average recognition rate is 90.41%. Experiments show that the method of sEMG signal and motion information fusion has good practicability in SLR.

Corticomuscular connectivity plays an important role in the neural control of human motion. This study recorded electroencephalography (EEG) and surface electromyography (sEMG) signals from subjects performing specific tasks (walking on level ground and on stairs) based on metronome instructions. This study presents a novel method based on vine copula to jointly model EEG and sEMG signals. The advantage of vine copula is its applicability in the construction of dependency structures to describe the connectivity between the cortex and muscles during different movements. A corticomuscular function network was also constructed by analyzing the dependence of each channel sample. The successfully constructed network shows information transmission between different divisions of the cortex, between muscles, and between the cortex and muscles when the body performs lower limb movements. Additionally, it highlights the potential of the vine copula concept used in this study, indicating that significant changes in the corticomuscular network under lower limb movements can be quantified by effective connectivity values.

The acquisition of good surface electromyography (sEMG) is an important prerequisite for correct and timely control of prosthetic limb movements. sEMG is nonlinear, nonstationary, and vulnerable against noise and a new sEMG denoising method using ensemble empirical mode decomposition (EEMD) and wavelet threshold is hence proposed to remove the random noise from the sEMG signal. With this method, the noised sEMG signal is first decomposed into several intrinsic mode functions (IMFs) by EEMD. The first IMF is mostly noise, coupled with a small useful component which is extracted using a wavelet transform based method by defining a peak-to-sum ratio and a noise-independent extracting threshold function. Other IMFs are processed using an improved wavelet threshold denoising method, where a noise variance estimation algorithm and an improved wavelet threshold function are combined. Key to the threshold denoising method, a threshold function is used to retain the required wavelet coefficients. Our denoising algorithm is tested for different sEMG signals produced by different muscles and motions. Experimental results show that the proposed new method performs better than other methods including the conventional wavelet threshold method and the EMD method, which guaranteed its usability in prosthetic limb control.

Major depressive disorder (MDD) is a common mental illness. This study used electroencephalography (EEG) to explore the effects of music therapy on brain networks in MDD patients and to elucidate changes in functional brain connectivity in subjects before and after musical stimulation. EEG signals were collected from eight MDD patients and eight healthy controls. The phase locking value was adopted to calculate the EEG correlation of different channels in different frequency bands. Correlation matrices and network topologies were studied to analyze changes in functional connectivity between brain regions. The results of the experimental analysis found that the connectivity of the delta and beta bands decreased, while the connectivity of the alpha band increased. Regarding the characteristics of the EEG functional network, the average clustering coefficient, characteristic path length and degree of each node in the delta band decreased significantly after musical stimulation, while the characteristic path length in the beta band increased significantly. Characterized by the average clustering coefficient and characteristic path length, the classification of depression and healthy controls reached 93.75% using a support vector machine.

In this paper, we propose a regression model to establish a relationship between surface electromyography (sEMG) and knee joint angles. In this model, the correlation dimension of wavelet coefficient (WCCD) and an Elman network are developed for the model estimation. In our experiment, the sEMG signals were recorded from five muscles concerned with knee joint motion, and knee joint angles were simultaneously recorded by a Codamotion system. First, we used a feature extraction method based on WCCD to extract optimal feature vectors from multichannel sEMG signals. Then, the Elman network was used to map the optimal sEMG features to the knee joint angles. The results show that the features extracted from the multichannel sEMG signals using the WCCD method proposed in this paper outperform the time-domain and frequency-domain methods. Our method is expected to be applied to intelligent prosthetics, exoskeleton robots, and medical rehabilitation robots.

The use of electroencephalography to recognize human emotions is a key technology for advancing human–computer interactions. This study proposes an improved deep convolutional neural network model for emotion classification using a non-end-to-end training method that combines bottom-, middle-, and top-layer convolution features. Four sets of experiments using 4500 samples were conducted to verify model performance. Simultaneously, feature visualization technology was used to extract the three-layer features obtained by the model, and a scatterplot analysis was performed. The proposed model achieved a very high accuracy of 93.7%, and the extracted features exhibited the best separability among the tested models. We found that adding redundant layers did not improve model performance, and removing the data of specific channels did not significantly reduce the classification effect of the model. These results indicate that the proposed model allows for emotion recognition with a higher accuracy and speed than the previously reported models. We believe that our approach can be implemented in various applications that require the quick and accurate identification of human emotions.

Corticomuscular functional coupling reflects the neuronal communication between cortical oscillations and muscle activity. Although the motor cortex is significantly involved in complex motor tasks, there is still no detailed understanding of the cortical contribution during such tasks. In this paper, we first propose a vine copula model to describe corticomuscular functional coupling and we construct the brain muscle function network. First, we recorded surface electromyography (sEMG) and electroencephalography (EEG) signals corresponding to the hand open, hand close, wrist flexion, and wrist extension motions of 12 participants during the initial experiments. The pre-processed signals were translated into the marginal density functions of different channels through the generalized autoregressive conditional heteroscedasticity model. Subsequently, we calculated the Kendall rank correlation coefficient, and used the R-vine model to decompose the multi-dimensional marginal density function into two-dimensional copula coefficient to determine the structure of the R-vine. Finally, we used the normalized adjacency matrix to structure the corticomuscular network for each hand motion considered. Based on the adjacency matrix, we found that the Kendall rank correlation coefficient between EEG and EMG was low. Moreover, a significant difference was observed in the correlation between the C3 and EMG signals for the different hand-motion activities. We also observed two core nodes in the networks corresponding to the four activities when the vine copula model was applied. Moreover, there was a large difference in the connections of the network models corresponding to the different hand-motion activities. Therefore, we believe that our approach is sufficiently accurate in identifying and classifying motor tasks.

Continuous joint angle estimation plays an important role in motion intention recognition and rehabilitation training. In this study, a surface electromyography- (sEMG-) mechanomyography (MMG) state-space model is proposed to estimate continuous multijoint movements from sEMG and MMG signals accurately. The model combines forward dynamics with a Hill-based muscle model that estimates joint torque only in a nonfeedback form, making the extended model capable of predicting the multijoint motion directly. The sEMG and MMG features, including the Wilson amplitude and permutation entropy, are then extracted to construct a measurement equation to reduce system error and external disturbances. Using the proposed model, a closed-loop prediction-correction approach, unscented particle filtering, is used to estimate the joint angle from sEMG and MMG signals. Comprehensive experiments are conducted on the human elbow and shoulder joint, and remarkable improvements are demonstrated compared with conventional methods.

Falls among the elderly comprise a major health problem. Daily activity monitoring and fall detection using wearable sensors provide an important healthcare system for elderly or frail individuals. We investigated the classification accuracy of daily activity and fall data based on surface electromyography (sEMG) and plantar pressure signals. sEMG and plantar pressure signals were collected, and their features were extracted. Suitable features were selected and combined for posture transition, gait, and fall using the Fisher class separability index. A feature-level fusion method, named as the global canonical correlation analysis of weighting genetic algorithm, was proposed to reduce dimensions. For the problem in which the number of daily activities is considerably more than the number of fall activities, Weighted Kernel Fisher Linear Discriminant Analysis (WKFDA) was proposed to classify gait and fall. Double Parameter Kernel Optimization based on Extreme Learning Machine (DPK-OMELM) was used to classify activities. Results showed that the classification accuracy of the posture transition is 100%, and the accuracy of gait and fall classified using WKFDA can reach 98%. For all types of posture transition, gait, and fall, sensitivity, specificity, and accuracy are over 96%.

Surface electromyography (sEMG) signals are commonly used in activity monitoring and rehabilitation training as they reflect effectively the motor intentions of users. This study proposed a new sEMG-based multifeature extraction and predictive model to predict knee-joint angle from multichannel sEMG. Six channels of sEMG from relevant muscles were recorded, and knee-joint angles were sampled simultaneously for six kinds of knee-joint movement models. The root–mean–square (RMS), wavelet coefficients (WC), and permutation entropy (PE) as features of sEMG were extracted. The back propagation neural network, generalized regression neural network, and least-square support vector regression machine (LS-SVR) were used as predictive models. To validate the effectiveness of the sEMG features and predictive models, twelve subjects without neural or musculoskeletal deficits participated in the experiment. Six kinds of knee-joint movement models at different speeds and different loads were respectively conducted by the subjects. Results revealed that the combination of the three features (RMS, WC, and PE) and LS-SVR performed well for the knee-joint-angle of all kinds of leg motions. The RMS error for all kinds of leg motions was

As an important research direction of human–computer interaction technology, gesture recognition is the key to realizing sign language translation. To improve the accuracy of gesture recognition, a new gesture recognition method based on four channel surface electromyography (sEMG) signals is proposed. First, the S-transform is applied to four channel sEMG signals to enhance the time-frequency detail characteristics of the signals. Then, multiscale singular value decomposition is applied to the multiple time-frequency matrix output of S-transform to obtain the time-frequency joint features with better robustness. The corresponding singular value permutation entropy is calculated as the eigenvalue to effectively reduce the dimension of multiple eigenvectors. The gesture features are used as input into the deep belief network for classification, and nine kinds of gestures are recognized with an average accuracy of 93.33%. Experimental results show that the multiscale singular value permutation entropy feature is especially suitable for the pattern classification of the deep belief network.

Surface electromyography (sEMG) signals are commonly used in activity monitoring and rehabilitation applications as they reflect effectively the motor intentions of users. However, real-time sEMG signals are non-stationary and vary to a large extent within the time frame of signals. Although previous studies have focused on the issues, their results have not been satisfactory. Therefore, we present a new method of conducting feature-level fusion to obtain a new feature space for sEMG signals. Eight activities of daily life (ADLs), including falls, were performed to obtain raw data from EMG signals from the lower limb. A feature set combining the time domain, time–frequency domain, and entropy domain was applied to the raw data to establish an initial feature space. A new projection method, the weighting genetic algorithm for GCCA (WGA-GCCA), was introduced to obtain the final feature space. Different tests were carried out to evaluate the performance of the new feature space. The new feature space created with the WGA-GCCA effectively reduced the dimensions and selected the best feature vectors dynamically while improving monotonicity. The Davies–Bouldin index (DBI) based on fuzzy c-means algorithms of the space obtained the lowest value compared with several fusion methods. It also achieved the highest accuracy when applied to support vector machine classifier.

As an essential subfield of context awareness, activity awareness, especially daily activity monitoring and fall detection, plays a significant role for elderly or frail people who need assistance in their daily activities. This study investigates the feature extraction and pattern recognition of surface electromyography (sEMG), with the purpose of determining the best features and classifiers of sEMG for daily living activities monitoring and fall detection. This is done by a serial of experiments. In the experiments, four channels of sEMG signal from wireless, wearable sensors located on lower limbs are recorded from three subjects while they perform seven activities of daily living (ADL). A simulated trip fall scenario is also considered with a custom-made device attached to the ankle. With this experimental setting, 15 feature extraction methods of sEMG, including time, frequency, time/frequency domain and entropy, are analyzed based on class separability and calculation complexity, and five classification methods, each with 15 features, are estimated with respect to the accuracy rate of recognition and calculation complexity for activity monitoring and fall detection. It is shown that a high accuracy rate of recognition and a minimal calculation time for daily activity monitoring and fall detection can be achieved in the current experimental setting. Specifically, the Wilson Amplitude (WAMP) feature performs the best, and the classifier Gaussian Kernel Support Vector Machine (GK-SVM) with Permutation Entropy (PE) or WAMP results in the highest accuracy for activity monitoring with recognition rates of 97.35% and 96.43%. For fall detection, the classifier Fuzzy Min-Max Neural Network (FMMNN) has the best sensitivity and specificity at the cost of the longest calculation time, while the classifier Gaussian Kernel Fisher Linear Discriminant Analysis (GK-FDA) with the feature WAMP guarantees a high sensitivity (98.70%) and specificity (98.59%) with a short calculation time (65.586 ms), making it a possible choice for pre-impact fall detection. The thorough quantitative comparison of the features and classifiers in this study supports the feasibility of a wireless, wearable sEMG sensor system for automatic activity monitoring and fall detection.