In this paper, we present a novel muscle synergy extraction method based on multivariate curve resolution–alternating least squares (MCR-ALS) to overcome the limitation of the nonnegative matrix factorization (NMF) method for extracting non-sparse muscle synergy, and we study its potential application for evaluating motor function of stroke survivors. Nonnegative matrix factorization (NMF) is the most widely used method for muscle synergy extraction. However, NMF is susceptible to components’ sparseness and usually provides inferior reliability, which significantly limits the promotion of muscle synergy. In this study, MCR-ALS was employed to extract muscle synergy from electromyography (EMG) data. Its performance was compared with two other matrix factorization algorithms, NMF and self-modeling mixture analysis (SMMA). Simulated data sets were utilized to explore the influences of the sparseness and noise on the extracted synergies. As a result, the synergies estimated by MCR-ALS were the most similar to true synergies as compared with SMMA and NMF. MCR-ALS was used to analyze the muscle synergy characteristics of upper limb movements performed by healthy (n = 11) and stroke (n = 5) subjects. The repeatability and intra-subject consistency were used to evaluate the performance of MCR-ALS. As a result, MCR-ALS provided much higher repeatability and intra-subject consistency as compared with NMF, which were important for the reliability of the motor function evaluation. The stroke subjects had lower intra-subject consistency and seemingly had more synergies as compared with the healthy subjects. Thus, MCR-ALS is a promising muscle synergy analysis method for motor function evaluation of stroke patients.

In the research of motor imagery brain‐computer interface (MI‐BCI), traditional electroencephalogram (EEG) signal recognition algorithms appear to be inefficient in extracting EEG signal features and improving classification accuracy. In this paper, we discuss a solution to this problem based on a novel step‐by‐step method of feature extraction and pattern classification for multiclass MI‐EEG signals. First, the training data from all subjects is merged and enlarged through autoencoder to meet the need for massive amounts of data while reducing the bad effect on signal recognition because of randomness, instability, and individual variability of EEG data. Second, an end‐to‐end sharing structure with attention‐based time‐incremental shallow convolution neural network is proposed. Shallow convolution neural network (SCNN) and bidirectional long short‐term memory (BiLSTM) network are used to extract frequency‐spatial domain features and time‐series features of EEG signals, respectively. Then, the attention model is introduced into the feature fusion layer to dynamically weight these extracted temporal‐frequency‐spatial domain features, which greatly contributes to the reduction of feature redundancy and the improvement of classification accuracy. At last, validation tests using BCI Competition IV 2a data sets show that classification accuracy and kappa coefficient have reached 82.7 ± 5.57% and 0.78 ± 0.074, which can strongly prove its advantages in improving classification accuracy and reducing individual difference among different subjects from the same network.

In motor imagery (MI) brain–computer interface (BCI) research, some researchers have designed MI paradigms of force under a unilateral upper-limb static state. It is difficult to apply these paradigms to the dynamic force interaction process between the robot and the patient in a brain-controlled rehabilitation robot system, which needs to induce thinking states of the patient’s demand for assistance. Therefore, in our research, according to the movement of wiping the table in human daily life, we designed a three-level-force MI paradigm under a unilateral upper-limb dynamic state. Based on the event-related de-synchronization (ERD) feature analysis of the electroencephalography (EEG) signals generated by the brain’s force change motor imagination, we proposed a multi-scale temporal convolutional network with attention mechanism (MSTCN-AM) algorithm to recognize ERD features of MI-EEG signals. Aiming at the slight feature differences of single-trial MI-EEG signals among different levels of force, the MSTCN module was designed to extract fine-grained features of different dimensions in the time–frequency domain. The spatial convolution module was then used to learn the area differences of space domain features. Finally, the attention mechanism dynamically weighted the time–frequency–space domain features to improve the algorithm’s sensitivity. The results showed that the accuracy of the algorithm was 86.4 ± 14.0% for the three-level-force MI-EEG data collected experimentally. Compared with the baseline algorithms (OVR-CSP+SVM (77.6 ± 14.5%), Deep ConvNet (75.3 ± 12.3%), Shallow ConvNet (77.6 ± 11.8%), EEGNet (82.3 ± 13.8%), and SCNN-BiLSTM (69.1 ± 16.8%)), our algorithm had higher classification accuracy with significant differences and better fitting performance.

Visuospatial selective attention can focus attention on a certain spatial area and rationally allocate attentional resources during visual target perception. Previous studies investigated the behavioral advantages of subjects when the target appeared in the upper and lower visual fields. However, the neurophysiological characteristics of the brain are not clear, and there is a lack of comprehensive analysis of the external behavior and the internal neurophysiological characteristics. We designed two task paradigms containing a spatial location orientation task and a visual search task. We used event-related potentials (ERP) components (P1 and P2) and electroencephalogram (EEG) rhythms (theta and alpha) to analyze the attention level and allocation of attention resources of the brain. The results showed that when the target appeared in the lower visual field in the spatial location orientation task, subjects consumed fewer attention resources and demonstrated better behavioral performance. In the visual search task, when the target appeared in the upper left visual field, subjects could better mobilize attention resources and behaved more advantageously. The study provides a basis for the design of the target in the upper and lower visual fields in the rehabilitation task, especially for stroke patients with low attention levels due to attention disorders such as spatial attention deficit.

Wearable gait analysis technology offering flexibility and cost-effectiveness, holds significant potential for quantifying the gait functions of stroke patients. It remains a key challenge that how to further reduce cumulative errors by using wearable technology. An algorithm for re-segmenting the gait phase was proposed to improve the precision of zero-velocity detection and gait parameters. The adaptive threshold method was used to separate the stationary phase and the non-stationary phase of the gait cycle. Next, an efficient method called the peak width threshold (PWT) for phase re-segmentation was developed to obtain effective peaks from acceleration and angular velocity data. And the peaks which were incorrectly segmented would be adjusted between the stationary and non-stationary phase. The precision of feet position detection was improved by using zero velocity update (ZUPT) algorithm and dual-foot constraints. Finally, four typical gait parameters were calculated from feet position and phase information. The results demonstrated that the wearable inertial sensors system displays a good agreement compared to the foot pressure-based gait analysis system, and the mean absolute errors (MAE) of the stride length, step length, stride velocity, and cadence are 1.77±1.22 cm, 2.40±1.83 cm, 0.02±0.01 m/s, and 1.16±1.37 steps/min, respectively. This proposed method improves the accuracy of gait phase and feet position detection, which is helpful to objectively evaluate the gait functions of stroke patients.