The implementation of surface electromyography (sEMG)-based hand gesture recognition on mobile and wearable systems is frequently restricted by the finite computing, memory, and battery capabilities of edge devices. Even though a low-density sEMG setup is a feasible hardware implementation, achieving robust recognition under such constraint conditions becomes very challenging due to the non-stationary nature and inter-subject variance. In this paper, we propose DSCAttenEMG, an efficient neural network that combines both Depthwise Separable Convolution (DSC) for local feature extraction and Multi-Head Self-Attention (MHSA) to model long-range dependencies on EMG/IMU data, using 1× 1 DSC followed by Global Average Pooling to replace high-dimensional fully connected layers. Extensive experimentation on a self-collected dataset, the public SeNic and BandMyo datasets shows that our approach achieves state-of-the-art recognition performance (94.45%, 94.11% and 92.89%) at negligible complexity (only 178–179K parameters). The model is capable of real-time inference (0.93ms on RTX 4090 GPU, 6.68ms on NVIDIA Jetson AGX Orin, 1.4/0.7ms on CPU/NPU of Qualcomm mobile platform) and has a high degree of practicality for embedded deployment (118 samples/s at$\sim$1 W on K230 edge AI platform). This amalgamation of three pivotal strengths, elevated accuracy, enhanced efficiency, and pragmatic viability, highlights its substantial potential for practical mobile and wearable applications.

To develop an efficient brain-computer interface (BCI) system, electroencephalography (EEG) measures neuronal activities in different brain regions through electrodes. Many EEG-based motor imagery (MI) studies do not make full use of brain network topology. In this paper, a deep learning framework based on a modified graph convolution neural network (M-GCN) is proposed, in which temporal-frequency processing is performed on the data through modified S-transform (MST) to improve the decoding performance of original EEG signals in different types of MI recognition. MST can be matched with the spatial position relationship of the electrodes. This method fusions multiple features in the temporal-frequency-spatial domain to further improve the recognition performance. By detecting the brain function characteristics of each specific rhythm, EEG generated by imaginary movement can be effectively analyzed to obtain the subjects' intention. Finally, the EEG signals of patients with spinal cord injury (SCI) are used to establish a correlation matrix containing EEG channel information, the M-GCN is employed to decode relation features. The proposed M-GCN framework has better performance than other existing methods. The accuracy of classifying and identifying MI tasks through the M-GCN method can reach 87.456%. After 10-fold cross-validation, the average accuracy rate is 87.442%, which verifies the reliability and stability of the proposed algorithm. Furthermore, the method provides effective rehabilitation training for patients with SCI to partially restore motor function.

Combing brain-computer interfaces (BCI) and virtual reality (VR) is a novel technique in the field of medical rehabilitation and game entertainment. However, the limitations of BCI such as a limited number of action commands and low accuracy hinder the widespread use of BCI-VR. Recent studies have used hybrid BCIs that combine multiple BCI paradigms and/or the multi-modal biosensors to alleviate these issues, which may become the mainstream of BCIs in the future. The main purpose of this review is to discuss the current status of multi-modal BCI-VR. This study first reviewed the development of the BCI-VR, and explored the advantages and disadvantages of incorporating eye tracking, motor capture, and myoelectric sensing into the BCI-VR system. Then, this study discussed the development trend of the multi-modal BCI-VR, hoping to provide a pathway for further research in this field.