The common spatial pattern (CSP) is a very effective feature extraction method in motor imagery based brain computer interface (BCI), but its performance depends on the selection of the optimal frequency band. Although a lot of research works have been proposed to improve CSP, most of these works have the problems of large computation costs and long feature extraction time. To this end, three new feature extraction methods based on CSP and a new feature selection method based on non-convex log regularization are proposed in this paper. Firstly, EEG signals are spatially filtered by CSP, and then three new feature extraction methods are proposed. We called them CSP-wavelet, CSP-WPD and CSP-FB, respectively. For CSP-Wavelet and CSP-WPD, the discrete wavelet transform (DWT) or wavelet packet decomposition (WPD) is used to decompose the spatially filtered signals, and then the energy and standard deviation of the wavelet coefficients are extracted as features. For CSP-FB, the spatially filtered signals are filtered into multiple bands by a filter bank (FB), and then the logarithm of variances of each band are extracted as features. Secondly, a sparse optimization method regularized with a non-convex log function is proposed for the feature selection, which we called LOG, and an optimization algorithm for LOG is given. Finally, ensemble learning is used for secondary feature selection and classification model construction. Combing feature extraction and feature selection methods, a total of three new EEG decoding methods are obtained, namely CSP-Wavelet+LOG, CSP-WPD+LOG, and CSP-FB+LOG. Four public motor imagery datasets are used to verify the performance of the proposed methods. Compared to existing methods, the proposed methods achieved the highest average classification accuracy of 88.86, 83.40, 81.53, and 80.83 in datasets 1–4, respectively. The feature extraction time of CSP-FB is the shortest. The experimental results show that the proposed methods can effectively improve the classification accuracy and reduce the feature extraction time. With comprehensive consideration of classification accuracy and feature extraction time, CSP-FB+LOG has the best performance and can be used for the real-time BCI system.

Feature extraction and selection are important parts of motor imagery electroencephalogram (EEG) decoding and have always been the focus and difficulty of brain-computer interface (BCI) system research. In order to improve the accuracy of EEG decoding and reduce model training time, new feature extraction and selection methods are proposed in this paper. First, a new spatial-frequency feature extraction method is proposed. The original EEG signal is preprocessed, and then the common spatial pattern (CSP) is used for spatial filtering and dimensionality reduction. Finally, the filter bank method is used to decompose the spatially filtered signals into multiple frequency subbands, and the logarithmic band power feature of each frequency subband is extracted. Second, to select the subject-specific spatial-frequency features, a hybrid feature selection method based on the Fisher score and support vector machine (SVM) is proposed. The Fisher score of each feature is calculated, then a series of threshold parameters are set to generate different feature subsets, and finally, SVM and cross-validation are used to select the optimal feature subset. The effectiveness of the proposed method is validated using two sets of publicly available BCI competition data and a set of self-collected data. The total average accuracy of the three data sets achieved by the proposed method is 82.39%, which is 2.99% higher than the CSP method. The experimental results show that the proposed method has a better classification effect than the existing methods, and at the same time, feature extraction and feature selection time also have greater advantages.

Diabetic patients need to test their blood glucose levels (BGL) frequently; however, traditional methods of blood collection and testing cause great pain to patients. In order to improve the quality of life of patients, this paper develops a noninvasive, portable, and continuous monitoring blood glucose detection system, which uses the latest bioimpedance integrated circuit to obtain the bioimpedance spectrum (BIS) of the inner forearm of the human body. The obtained BIS covers most of the frequencies up to 1 MHz. A BGL estimation model is developed using sparse group least absolute shrinkage and selection operator combined with a Gaussian kernel function support vector regression to select the optimal frequencies and features for BIS. The correlations between different frequencies and features and BGL are investigated. We test our system on a collected dataset of clinical subjects, and the results show that the average mean absolute relative difference for all subjects is 9.90%, the root mean square error is 14.81 mg/dl, and the mean absolute error is 11.75 mg/dl. 100% of the estimates fall in zones A and B of the Clarke error grid. Preliminary results show that the use of BIS integrated circuits in combination with machine learning techniques promises to enable portable, noninvasive, continuous monitoring of BGLs.

Background/Objectives: Wearable affective human–computer interaction increasingly relies on sparse-channel EEG signals to ensure comfort and practicality in real-life scenarios. However, the limited information provided by sparse-channel EEG, together with pronounced inter-subject variability, makes reliable cross-subject emotion recognition particularly challenging. Methods: To address these challenges, we propose a cross-subject emotion recognition model, termed TSCL-LwF, based on sparse-channel EEG. It combines a multi-scale convolutional network (TSCL) and an incremental learning strategy with Learning without Forgetting (LwF). Specifically, the TSCL is utilized to capture the spatio-temporal characteristics of sparse-channel EEG, which employs diverse receptive fields of convolutional networks to extract and fuse the interaction information within the local prefrontal area. The incremental learning strategy with LwF introduces a limited set of labeled target domain data and incorporates the knowledge distillation loss to retain the source domain knowledge while enabling rapid target domain adaptation. Results: Experiments on the DEAP dataset show that the proposed TSCL-LwF achieves accuracy of 77.26% for valence classification and 80.12% for arousal classification. Moreover, it also exhibits superior accuracy when evaluated on the self-collected dataset EPPVR. Conclusions: The successful implementation of cross-subject emotion recognition based on a sparse-channel EEG will facilitate the development of wearable EEG technologies with practical applications.

Time–frequency–spatial (TFS) features play a crucial role in motor imagery electroencephalogram (EEG) classification. However, effectively leveraging these multidimensional features to enhance classification accuracy remains a significant challenge. Although feature selection techniques are widely used to extract informative TFS representations, most existing methods treat each dimension independently, overlooking the intrinsic grouping and hierarchical relationships among them. To address this limitation, this paper proposes a hierarchical feature recalibration network (HFRNet) that explicitly models intergroup dependencies and the hierarchical structure of TFS features, thereby substantially improving motor imagery EEG classification performance. HFRNet employs a two‐layer weighting mechanism for hierarchical feature recalibration, followed by a classification module. In the first weighting layer, spatial features within each time–frequency segment are grouped and represented as feature maps. Channel‐wise dependencies are captured through squeeze‐and‐excitation operations to learn channel weights, which are then used to rescale each feature map. In the second weighting layer, the recalibrated features are reorganized across time windows and further refined through a similar recalibration process. Finally, in the classification block, the refined features are flattened, concatenated into a single feature vector, and passed through dropout and fully connected (FC) layers for classification. Extensive experiments conducted on five motor imagery datasets demonstrate that the proposed HFRNet achieves the best overall performance, with an average accuracy (F1 score) of 81.03% (0.7931). Comparative evaluations against 30 feature selection methods and recent state‐of‐the‐art approaches further confirm the superior effectiveness and robustness of the proposed model.