Motor Imagery EEG Signal Classification Using Distinctive Feature Fusion with Adaptive Structural LASSO.

A motor imagery brain-computer interface connects the human brain and computers via electroencephalography (EEG). However, individual differences in the frequency ranges of brain activity during motor imagery tasks pose a challenge, limiting the manual feature extraction for motor imagery classification. To extract features that match specific subjects, we proposed a novel motor imagery classification model using distinctive feature fusion with adaptive structural LASSO. Specifically, we extracted spatial domain features from overlapping and multi-scale sub-bands of EEG signals and mined discriminative features by fusing the task relevance of features with spatial information into the adaptive LASSO-based feature selection. We evaluated the proposed model on public motor imagery EEG datasets, demonstrating that the model has excellent performance. Meanwhile, ablation studies and feature selection visualization of the proposed model further verified the great potential of EEG analysis.

Comprehensive Similarity Algorithm and Molecular Dynamics Simulation-Assisted Terahertz Spectroscopy for Intelligent Matching Identification of Quorum Signal Molecules (N-Acyl-Homoserine Lactones).

To investigate the mechanism of aquatic pathogens in quorum sensing (QS) and decode the signal transmission of aquatic Gram-negative pathogens, this paper proposes a novel method for the intelligent matching identification of eight quorum signaling molecules (N-acyl-homoserine lactones, AHLs) with similar molecular structures, using terahertz (THz) spectroscopy combined with molecular dynamics simulation and spectral similarity calculation. The THz fingerprint absorption spectral peaks of the eight AHLs were identified, attributed, and resolved using the density functional theory (DFT) for molecular dynamics simulation. To reduce the computational complexity of matching recognition, spectra with high peak matching values with the target were preliminarily selected, based on the peak position features of AHL samples. A comprehensive similarity calculation (CSC) method using a weighted improved Jaccard similarity algorithm (IJS) and discrete Fréchet distance algorithm (DFD) is proposed to calculate the similarity between the selected spectra and the targets, as well as to return the matching result with the highest accuracy. The results show that all AHL molecular types can be correctly identified, and the average quantization accuracy of CSC is 98.48%. This study provides a theoretical and data-supported foundation for the identification of AHLs, based on THz spectroscopy, and offers a new method for the high-throughput and automatic identification of AHLs.

Efficient Feature Learning Model of Motor Imagery EEG Signals with L1-Norm and Weighted Fusion.

Brain-computer interface (BCI) for motor imagery is an advanced technology used in the field of medical rehabilitation. However, due to the poor accuracy of electroencephalogram feature classification, BCI systems often misrecognize user commands. Although many state-of-the-art feature selection methods aim to enhance classification accuracy, they usually overlook the interrelationships between individual features, indirectly impacting the accuracy of feature classification. To overcome this issue, we propose an adaptive feature learning model that employs a Riemannian geometric approach to generate a feature matrix from electroencephalogram signals, serving as the model's input. By integrating the enhanced adaptive L1 penalty and weighted fusion penalty into the sparse learning model, we select the most informative features from the matrix. Specifically, we measure the importance of features using mutual information and introduce an adaptive weight construction strategy to penalize regression coefficients corresponding to each variable adaptively. Moreover, the weighted fusion penalty balances weight differences among correlated variables, reducing the model's overreliance on specific variables and enhancing accuracy. The performance of the proposed method was validated on BCI Competition IV datasets IIa and IIb using the support vector machine. Experimental results demonstrate the effectiveness and superiority of the proposed model compared to the existing models.

Application of Graphene-Combined Rare-Earth Oxide (Sm2O3) in Solar-Blind Ultraviolet Photodetection.

Rare-earth oxide Sm2O3 is theoretically expected to be used in the preparation of ultraviolet (UV) detectors with low dark currents and high radiation resistance due to its characteristics of a wide bandgap, a high dielectric constant, and high chemical stability. However, certain features that rare-earth oxides possess, such as high resistivity and weak photoelectric response currents, have hindered relevant research on these kinds of materials in the field of UV detection. In this work, a p-Gr/i-Sm2O3/n-SiC heterojunction photovoltaic solar-blind UV sensor was constructed for the first time. Because of the high mobility of graphene (Gr) and the contribution of double built-in electric fields in the heterojunction, the collection efficiency of photogenerated carriers has been greatly improved, with the typical shortcomings of high resistivity and poor photoelectric response performance of rare-earth oxides having been overcome. This detector has exhibited outstanding performance at 0 V, including a responsivity of 19.8 mA/W and an open-circuit voltage of 0.68 V. Additionally, this detector has a detectivity as high as 1.2 × 1011 jones, which is at the front position of most ultraviolet detectors. The fabrication of this high-performance Sm2O3-based photovoltaic UV detector has broadened the application fields of rare-earth oxide semiconductors. Therefore, this project has important value for future research in relevant fields.

Signal-BNF: a Bayesian network fusing approach to predict signal peptides.

A signal peptide is a short peptide chain that directs the transport of a protein and has become the crucial vehicle in finding new drugs or reprogramming cells for gene therapy. As the avalanche of new protein sequences generated in the postgenomic era, the challenge of identifying new signal sequences has become even more urgent and critical in biomedical engineering. In this paper, we propose a novel predictor called Signal-BNF to predict the N-terminal signal peptide as well as its cleavage site based on Bayesian reasoning network. Signal-BNF is formed by fusing the results of different Bayesian classifiers which used different feature datasets as its input through weighted voting system. Experiment results show that Signal-BNF is superior to the popular online predictors such as Signal-3L and PrediSi. Signal-BNF is featured by high prediction accuracy that may serve as a useful tool for further investigating many unclear details regarding the molecular mechanism of the zip code protein-sorting system in cells.

Automatic detection of abnormal EEG signals using multiscale features with ensemble learning.

Electroencephalogram (EEG) is an economical and convenient auxiliary test to aid in the diagnosis and analysis of brain-related neurological diseases. In recent years, machine learning has shown great potential in clinical EEG abnormality detection. However, existing methods usually fail to consider the issue of feature redundancy when extracting the relevant EEG features. In addition, the importance of utilizing the patient age information in EEG detection is ignored. In this paper, a new framework is proposed for distinguishing an unknown EEG recording as either normal or abnormal by identifying different types of EEG-derived significant features. In the proposed framework, different hierarchical salient features are extracted using a time-wise multi-scale aggregation strategy, based on a selected group of statistical characteristics calculated from the optimum discrete wavelet transform coefficients. We also fuse the age information with multi-scale features for further improving discrimination. The integrated features are classified using three ensemble learning classifiers, CatBoost, LightGBM, and random forest. Experimental results show that our method with CatBoost classifier can yield superior performance vis-a-vis competing techniques, which indicates the great promise of our methodology in EEG pathology detection.

Neuro-Behavioral Dynamic Prediction of Interpersonal Cooperation and Aggression.

How to quickly predict an individual's behavioral choices is an important issue in the field of human behavior research. Using noninvasive electroencephalography, we aimed to identify neural markers in the prior outcome-evaluation stage and the current option-assessment stage of the chicken game that predict an individual's behavioral choices in the subsequent decision-output stage. Hierarchical linear modeling-based brain-behavior association analyses revealed that midfrontal theta oscillation in the prior outcome-evaluation stage positively predicted subsequent aggressive choices; also, beta oscillation in the current option-assessment stage positively predicted subsequent cooperative choices. These findings provide electrophysiological evidence for the three-stage theory of decision-making and strengthen the feasibility of predicting an individual's behavioral choices using neural oscillations.

Transcription factors GATA4 and TBX5 promote cardiomyogenic differentiation of rat bone marrow mesenchymal stromal cells.

Bone marrow mesenchymal stromal cells (BMSCs) have potential applications in cell and gene therapies for cardiac disease. The cardiac-specific transcription factors GATA-binding protein 4 (GATA4) and T-Box protein 5 (TBX5) are considered to be pivotal in cardiogenesis. The aim of this study was to investigate the effects of GATA4 and TBX5 on cardiomyogenic differentiation of BMSCs. The BMSCs were initially isolated and identified. Vectors harboring cardiac transcription factor genes GATA4 and TBX5 or empty vectors were transferred into BMSCs. Cardiomyogenic cells differentiated from BMSCs were identified by expression of cardiac-specific markers including cardiac troponin T, connexin 43, ß-myosin heavy chain, and myosin light chain-2 using immunocytochemical staining, western blotting, and quantitative real-time PCR. The ultrastructures of the differentiated cells were examined by transmission electron microscopy, which were similar to those of fetal cardiomyocytes. The differentiated cells exhibited L-type calcium current activities reflective of the electrophysiological characteristics of cardiomyocytes. These findings indicate that exogenous expression of cardiac-specific transcription factors GATA4 and TBX5 enhance cardiomyogenic differentiation of BMSCs.