Comprehensive analysis on subchondral bone marrow lesions of human osteoarthritis by integrating bulk and single-cell transcriptomes.

OBJECTIVE: This study aims to demonstrate the cellular composition and underlying mechanisms in subchondral bone marrow lesions (BMLs) of knee osteoarthritis (OA). METHODS: BMLs were assessed by MRI Osteoarthritis Knee Score (MOAKS)≥2. Bulk RNA-sequencing (bulk-seq) and BML-specific differentially expressed genes (DEGs) analysis were performed among subchondral bone samples (including OA-BML=3, paired OA-NBML=3; non-OA=3). The hub genes of BMLs were identified by verifying in independent datasets and multiple bioinformatic analyses. To further estimate cell-type composition of subchondral bone, we utilized two newly developed deconvolution algorithms (MuSiC, MCP-counter) in transcriptomic datasets, based on signatures from open-accessed single-cell RNA sequencing (scRNA-seq). Finally, competing endogenous RNA (ceRNA) and transcription factor (TF) networks were constructed through multiple predictive databases, and validated by public non-coding RNA profiles. RESULTS: A total of 86 BML-specific DEGs (up 79, down 7) were identified. IL11 and VCAN were identified as core hub genes. The "has-miR-424-5p/lncRNA PVT1" was determined as crucial network, targeting IL11 and VCAN, respectively. More importantly, two deconvolution algorithms produced approximate estimations of cell-type composition, and the cluster of heterotopic-chondrocyte was discovered abundant in BMLs, and positively correlated with the expression of hub genes. CONCLUSION: IL11 and VCAN were identified as the core hub genes of BMLs, and their molecular networks were determined as well. We profiled the characteristics of subchondral bone at single-cell level and determined that the heterotopic-chondrocyte was abundant in BMLs and was closely linked to IL11 and VCAN. Our study may provide new insights into the microenvironment and pathological molecular mechanism of BMLs, and could lead to novel therapeutic strategies.

Asporin regulated by miR-26b-5p mediates chondrocyte senescence and exacerbates osteoarthritis progression via TGF-ß1/Smad2 pathway.

OBJECTIVES: This study aimed to investigate the role and mechanism of asporin in modulating chondrocyte senescence in OA pathology. METHODS: Asporin and senescence-related hallmark expression were examined in human and experimental OA mouse cartilage samples. Twelve-week-old male C57 mice were administered with recombinant protein (rm-asporin)- or asporin-siRNA-expressing lentiviruses via intra-articular injection once a week after destabilization of the medial meniscus (DMM) surgery to induce OA. Cartilage damage was measured using the Osteoarthritis Research Society International score. Senescence-associated ß-galactosidase (SA-ß-Gal) staining, γH2AX, p21 and p16INK4a were analysed by immunofluorescence staining and western blot to assess the specific role of asporin in chondrocyte senescence. The TGF-ß1-Smad2 signalling pathway and miR-26b-5p were further evaluated to explore the mechanism of asporin in OA. RESULTS: Asporin was upregulated in articular chondrocytes of OA patients and DMM mice and accompanied by accumulation of senescent cells. Asporin overexpression exaggerated OA progression, whereas silencing asporin restored chondrocyte homeostasis and deferred chondrocyte senescence, leading to markedly attenuated DMM-induced OA. Cellular and molecular analyses showed that asporin can be inhibited by miR-26b-5p, which was significantly downregulated in OA cartilage, leading to exacerbation of experimental OA partially through inhibition of TGF-ß1-Smad2 signalling in chondrocytes. CONCLUSIONS: Our findings indicate that asporin plays an essential role in chondrocyte senescence and OA pathogenesis. Upregulated by miR-26b-5p, asporin inhibits the TGF-ß1-Smad2 pathway to accelerate chondrocyte senescence and exacerbate cartilage degeneration. Targeting the miR-26b-5p-asporin-Smad2 axis may serve as a practical therapeutic strategy to delay chondrocyte senescence and OA development.

Long non-coding RNA MEG3 silencing and microRNA-214 restoration elevate osteoprotegerin expression to ameliorate osteoporosis by limiting TXNIP.

Studies have shown that long non-coding RNA (lncRNA) MEG3 plays a key role in osteoporosis (OP), but its regulatory mechanism is somewhat incompletely clear. Here, we intend to probe into the mechanism of MEG3 on OP development by modulating microRNA-214 (miR-214) and thioredoxin-interacting protein (TXNIP). Rat models of OP were established. MEG3, miR-214 and TXNIP mRNA expression in rat femoral tissues were detected, along with TXNIP, OPG and RANKL protein expression. BMD, BV/TV, Tb.N and Tb.Th in tissue samples were measured. Ca, P and ALP contents in rat serum were also determined. Primary osteoblasts were isolated and cultured. Viability, COL-I, COL-II and COL-Χ mRNA expression, PCNA, cyclin D1, OCN, RUNX2 and osteolix protein expresion, ALP content and activity, and mineralized nodule area of rat osteoblasts were further detected. Dual-luciferase reporter gene and RNA-pull down assays verified the targeting relationship between MEG3, miR-214 and TXNIP. MEG3 and TXNIP were up-regulated while miR-214 was down-regulated in femoral tissues of OP rats. MEG3 silencing and miR-214 overexpression increased BMD, BV/TV, Tb.N, Tb.Th, trabecular bone area, collagen area and OPG expression, and down-regulated RANKL of femoral tissues in OP rats. MEG3 silencing and miR-214 overexpression elevated Ca and P and reduced ALP in OP rat serum, elevated osteoblast viability, differentiation ability, COL-I and COL-Χ expression and ALP activity, and reduced COL-II expression of osteoblasts. MEG3 specifically bound to miR-214 to regulate TXNIP. MEG3 silencing and miR-214 overexpression promote proliferation and differentiation of osteoblasts in OP by down-regulating TXNIP, which further improves OP.

MicroRNA-497 elevation or LRG1 knockdown promotes osteoblast proliferation and collagen synthesis in osteoporosis via TGF-ß1/Smads signalling pathway.

MicroRNAs (miRNAs) have been corroborated to engage in the process of cellular activities in osteoporosis. However, few researches have been conducted to expose the integrated role of miR-497, leucine-rich alpha-2-glycoprotein-1 (LRG1) and transforming growth factor beta 1 (TGF-ß1)/Smads signalling pathway in osteoporosis. Thereafter, the study is set out to delve into miR-497/LRG1/TGF-ß1/Smads signalling pathway axis in osteoporosis. Osteoporosis bone tissues and normal bone tissues were collected. Rat osteoporosis models were constructed via ovariectomy. Model rats were injected with restored miR-497 or depleted LRG1 to explore their roles in osteoporosis. Rat osteoblasts were extracted from osteoporosis rats and transfected with restored miR-497 or depleted LRG1 for further verification. MiR-497 and LRG1 expression in femoral head tissues and osteoblasts of osteoporosis rats were detected. TGF-ß1/Smads signalling pathway-related factors were detected. MiR-497 was poorly expressed while LRG1 was highly expressed and TGF-ß1/Smads signalling pathway activation was inhibited in osteoporosis. MiR-497 up-regulation or LRG1 down-regulation activated TGF-ß1/Smads signalling pathway, promoted collagen type 1 synthesis and suppressed oxidative stress in femoral head tissues in osteoporosis. MiR-497 restoration or LRG1 knockdown activated TGF-ß1/Smads signalling pathway, promoted viability and suppressed apoptosis of osteoblasts in osteoporosis. Our study suggests that miR-497 up-regulation or LRG1 down-regulation promotes osteoblast viability and collagen synthesis via activating TGF-ß1/Smads signalling pathway, which may provide a novel reference for osteoporosis treatment.

The Coupling of Taylor Dispersion Analysis and Mass Spectrometry to Differentiate Protein Conformations.

Measuring the conformations of protein and protein-ligand complexes in solution is critical for investigating protein bioactivities, but their rapid analyses remain as challenging problems. Here, we report the coupling of Taylor dispersion analysis (TDA) with mass spectrometry (MS) for the rapid conformation differentiation of protein and noncovalent protein complex in solution environments. First, a branched capillary design was applied to achieve double band detection for the peak retention time correction in TDA measurements. After ionization, analytes were further detected and distinguished by their mass to charge (m/z) ratios in the consequent MS analysis. As a result, protein or protein complex in a mixture could be analyzed in terms of both hydrodynamic radius and m/z. The feasibility of this method was verified by analyzing a mixture of angiotensin II and phenylalanine, and the conformations of cytochrome C at different pH conditions were then investigated. As proof-of-concept demonstrations, the complexes of tri-N-acetylchitotriose with two proteins (lysozyme and cytochrome C) were characterized with results verified by molecular dynamics simulations. The TDA-MS method is promising for rapid structural analyses of trace amounts protein-ligand complexes, which could potentially be used to differentiate intact protein or protein complex conformations.

Recurrent arthritis caused by Candida parapsilosis: a case report and literature review.

BACKGROUND: Candida arthritis is extremely rare and also represents a major challenge of diagnosis and treatment. Here we reported a rare case of recurrent arthritis caused by Candida parapsilosis. CASE PRESENTATION: A 56-year-old Chinese male suffered from recurrent pain and swelling in his right knee after several times of "small needle-knife" acupuncture and corticosteroid injection of the joint. Candida parapsilosis was cultured in his synovial fluid and identified by sequencing of its Internal Transcribed Spacer (ITS) gene. Here we present the radiological characteristics, arthroscopic pictures, and synovium pathology of this patient. Also, blood test and chemical analysis of his synovial fluid were listed as well as the ITS sequence of this Candida species identified. The patient underwent thorough arthroscopic debridement and then set on fluconazole 400 mg daily for 12 months. His symptoms resolved and no relapse was observed on the last follow-up. Additionally, a brief but comprehensive review of C. parapsilosis arthritis episodes from past to now were studied. CONCLUSION: With the detailed clinical information reported in this case and our literature review, we hope they would add to our knowledge of C. parapsilosis arthritis - its clinical settings, laboratory features, radiological characteristics, arthroscopic findings and experience of management.

GSK-3ß activation index is a potential indicator for recurrent inflammation of chronic rhinosinusitis without nasal polyps.

Chronic rhinosinusitis without nasal polyps (CRSsNP) is one of the most common otorhinolaryngologic diseases worldwide. However, the underlying mechanism remains unclear. In this study, the expression of glycogen synthase kinase 3 (GSK-3) was quantitatively evaluated in patients with CRSsNP (n = 20) and healthy controls (n = 20). The mRNA levels of GSK-3α and GSK-3ß were examined by qPCR, the immunoreactivities of GSK-3ß and nuclear factor-κB (NF-κB) were examined by immunohistochemistry (IHC) staining, and the protein levels of GSK-3ß, phospho-GSK-3ß (p-GSK-3ß, s9) and NF-κB were examined using Western blot analysis. We found that GSK-3 was highly expressed in both CRSsNP and control groups without significant difference in both GSK-3ß mRNA and protein levels. However, when compared with healthy control group, the GSK-3ß activation index, defined as the ratio of GSK-3ß over p-GSK-3ß, was significantly decreased, whereas the NF-κB protein abundance was significantly increased in CRSsNP group (P < 0.05). Strikingly, the GSK-3ß activation index, was highly correlated with NF-κB protein level, as well as CT scores in CRSsNP group (P < 0.05). It was also highly correlated with the mRNA expressions of inflammation-related genes, including T-bet, IFN-γ and IL-4 in CRSsNP group (P < 0.05). Our findings suggest that GSK-3ß activation index, reflecting the inhibitory levels of GSK-3ß through phosphorylation, may be a potential indicator for recurrent inflammation of CRSsNP, and that the insufficient inhibitory phosphorylation of GSK-3ß may play a pivotal role in the pathogenesis of CRSsNP.

Effects of mobile phone WeChat services improve adherence to corticosteroid nasal spray treatment for chronic rhinosinusitis after functional endoscopic sinus surgery: a 3-month follow-up study.

The objective of the study is to investigate the impact of receiving daily WeChat services on one's cell phone on adherence to corticosteroid nasal spray treatment in chronic rhinosinusitis (CRS) patients after functional endoscopic sinus surgery (FESS). This study was a two-arm, randomized, follow-up investigation. Patients with chronic rhinosinusitis with/without nasal polyps following bilateral FESS were randomised to receive, or to not receive, daily WeChat service on their cell phone to take corticosteroid nasal spray treatment. A prescription of budesonide aqueous nasal spray 128 µg bid was given to all the subjects. Then they returned to the clinic after 30, 60, 90 days. The primary study outcome was adherence to nasal spray treatment, whereas secondary outcomes were change in endoscopic findings and SinoNasal Outcome Test-20 (SNOT-20). On the whole, there was a significant inter-group difference in the change of adherence rate (F = 90.88, p = 0.000). The WeChat group had much higher adherence rate than the control group during the follow-up. In terms of postoperative endoscopic scores and SNOT-20, except granulation score, no significant differences were observed between the two randomization groups. WeChat services are already after a short period of observation associated with improved adherence to corticosteroid nasal spray treatment in CRS patients after FESS.

Epigallocatechin Gallate (EGCG) Inhibits Alpha-Synuclein Aggregation: A Potential Agent for Parkinson's Disease.

Protein aggregation is a prominent feature of many neurodegenerative disorders including Parkinson's disease (PD). Aggregation of alpha-synuclein (SNCA) may underlie the pathology of PD. They are the main components of Lewy bodies and dystrophic neurites that are the intraneuronal inclusions characteristic of the disease. We have demonstrated that the polyphenol (-)-epi-gallocatechine gallate (EGCG) inhibited SNCA aggregation, which made it a candidate for therapeutic intervention in PD. Three methods were used: SNCA fibril formation inhibition by EGCG in incubates; inhibition of the SNCA fluorophore A-Syn-HiLyte488 binding to plated SNCA in microwells; and inhibition of the A-Syn-HiLyte488 probe binding to aggregated SNCA in postmortem PD tissue. Recombinant human SNCA was incubated under conditions that result in fibril formation. The aggregation was blocked by 100 nM EGCG in a concentration-dependent manner, as shown by an absence of thioflavin T binding. In the microplate assay system, the ED50 of EGCG inhibition of A-Syn-HiLyte488 binding to coated SNCA was 250 nM. In the PD tissue based assay, SNCA aggregates were recognized by incubation with 7 nM of A-Syn-HiLyte488. This binding was blocked by EGCG in a concentration dependent manner. The SNCA amino acid sites, which potentially interacted with EGCG, were detected on peptide membranes. It was implicated that EGCG binds to SNCA by instable hydrophobic interactions. In this study, we suggested that EGCG could be a potent remodeling agent of SNCA aggregates and a potential disease modifying drug for the treatment of PD and other α-synucleinopathies.

FTY-720P Suppresses Osteoclast Formation by Regulating Expression of Interleukin-6 (IL-6), Interleukin-4 (IL-4), and Matrix Metalloproteinase 2 (MMP-2).

BACKGROUND Osteoclast formation is closely related to the immune system. FTY720, a new immunosuppressive agent, has some functions in immune regulation. Its main active ingredients become FTY-720P in vivo by phosphorylation modification. The objective of this study was to determine the effects of FTY-720 with various concentrations on osteoclasts in vitro. MATERIAL AND METHODS RAW264.7 cells and bone marrow-derived mononuclear phagocytes (BMMs) were treated with RANKL to obtain osteoclasts in vitro. To investigate the role of FTY-720 in osteoclast formation, trap enzyme staining was performed and the number of osteoclasts was counted. Bone slices were stained with methylene blue, we counted the number of lacunae after bone slices were placed into dishes together with osteoclasts, and we observed the effect and function of FTY-720 in osteoclasts induced by RAW264.7 cells and BMMs. Then, we used a protein array kit to explore the effects of FTY-720P on osteoclasts. RESULTS The results of enzyme trap staining and F-actin staining experiments show that, with the increasing concentration of FTY-720P, the number of osteoclast induced by RAW264.7 cells and BMMs gradually decreased (P<0.05), especially when the FTY-720P concentration reached 1000 ng/ml, and the number of osteoclasts formed was the lowest (P<0.05). With bone lacuna toluidine blue staining, the results also show that, with the increasing concentration of FTY-720P, the number of bone lacuna gradually decreased (P<0.05), and the number of lacunae is lowest when the concentration reached 800 ng/ml. Finally, protein array results showed that IL-4, IL-6, IL-12, MMP-2, VEGF-C, GFR, basic FGF, MIP-2, and insulin proteins were regulated after FTY-720P treatment. CONCLUSIONS FTY-720P can suppress osteoclast formation and function, and FTY-720P induces a series of cytokine changes.

c-Jun N-terminal kinase - c-Jun pathway transactivates Bim to promote osteoarthritis.

Osteoarthritis (OA) is a chronic degenerative joint disorder. Previous studies have shown abnormally increased apoptosis of chondrocytes in patients and animal models of OA. TNF-α and nitric oxide have been reported to induce chondrocyte ageing; however, the mechanism of chondrocyte apoptosis induced by IL-1ß has remained unclear. The aim of this study is to identify the role of the c-Jun N-terminal kinase (JNK) - c-Jun pathway in regulating induction of Bim, and its implication in chondrocyte apoptosis. This study showed that Bim is upregulated in chondrocytes obtained from the articular cartilage of OA patients and in cultured mouse chondrocytes treated with IL-1ß. Upregulation of Bim was found to be critical for chondrocyte apoptosis induced by IL-1ß, as revealed by the genetic knockdown of Bim, wherein apoptosis was greatly reduced in the chondrocytes. Moreover, activation of the JNK-c-Jun pathway was observed under IL-1ß treatment, as indicated by the increased expression levels of c-Jun protein. Suppression of the JNK-c-Jun pathway, using chemical inhibitors and RNA interference, inhibited the Bim upregulation induced by IL-1ß. These findings suggest that the JNK-c-Jun pathway is involved in the upregulation of Bim during OA and that the JNK-c-Jun-Bim pathway is vital for chondrocyte apoptosis.

MST-net: A multi-scale swin transformer network for EEG-based cognitive load assessment.

Cognitive load assessment plays a crucial role in monitoring safe production, resource allocation, and subjective initiative in human-computer interaction. Due to its high time resolution and convenient acquisition, Electroencephalography (EEG) is widely applied in brain monitoring and cognitive state assessment. In this study, a multi-scale Swin Transformer network (MST-Net) was proposed for cognitive load assessment, which extracts local features with different sensory fields using a multi-scale parallel convolution model and introduces the attention mechanism of the Swin Transformer to obtain the feature correlations among multi-scale local features. The performance of the proposed network was validated using the EEG signals collected during cognitive tasks and N-back tasks with three different load levels. Results show that the MST-Net network achieved the best classification accuracy on both local and public datasets, and was higher than the mainstream Swin Transformer and CNN. Furthermore, results of ablation experiments and feature visualization revealed that the proposed MST-Net could well characterize different cognitive loads, which not only provided novel and powerful tools for cognitive load assessment but also showed potential for broad application in brain-computer interface (BCI) systems.

Identification of osteoblastic autophagy-related genes for predicting diagnostic markers in osteoarthritis.

The development of osteoarthritis (OA) involves subchondral bone lesions, but the role of osteoblastic autophagy-related genes (ARGs) in osteoarthritis is unclear. Through integrated analysis of single-cell dataset, Bulk RNA dataset, and 367 ARGs extracted from GeneCards, 40 ARGs were found. By employing multiple machine learning algorithms and PPI networks, three key genes (DDIT3, JUN, and VEGFA) were identified. Then the RF model constructed from these genes indicated great potential as a diagnostic tool. Furthermore, the model's effectiveness in predicting OA has been confirmed through external validation datasets. Moreover, the expression of ARGs was examined in osteoblasts subject to excessive mechanical stress, human and mouse tissues. Finally, the role of ARGs in OA was confirmed through co-culturing explants and osteoblasts. Thus, osteoblastic ARGs could be crucial in OA development, providing potential diagnostic and treatment strategies.

TCF12 regulates the TGF-ß/Smad2/3 signaling pathway to accelerate the progression of osteoarthritis by targeting CXCR4.

Objective: Osteoarthritis (OA), which involves total joint damage and dysfunction, is a leading cause of disability worldwide. However, its exact pathogenesis remains unclear. Here, we identified TCF12 as an important regulator of the progression of OA. Methods: qRT-PCR, immunoblotting and immunohistochemistry (IHC) were used to detect the expression level of TCF12. The interaction of TCF12 with its downstream factor CXCR4 was assessed by Western blotting, immunofluorescence, qRT-PCR and luciferase assays. A mouse model was generated to examine the functions and mechanism of TCF12 in vivo. Result: TCF12 expression was upregulated in chondrocytes stimulated with IL-1ß and osteoarthritic chondrocytes. TCF12 upregulates the expression of CXCR4 and leads to dysfunction of the TGF-ß signaling pathway. Furthermore, knockdown of TCF12 alleviated cartilage damage in a mouse model generated by destabilization of the medial meniscus (DMM). Conclusion: TCF12 aggravates the progression of OA by targeting CXCR4 and then activating the TGF-ß signaling pathway, suggesting that TCF12 may be a new target for the treatment of OA. The translational potential of this article: Transcription Factor 12(TCF12), is known to regulate cell development and differentiation, It has been widely studied in various organs and diseases, but its role in OA remains unclear. Here, we identified Transcription Factor 12(TCF12) as an important regulator mediating chondrocyte senescence and cartilage extracellular matrix degradation indicating its role in OA. We found that TCF12 expression was upregulated both locally and systemically as OA advanced in patients with OA, and in mice after DMM surgery to induce OA. TCF12 expression caused striking progressive articular cartilage damage, synovial hyperplasia in OA mice, and remarkably, it was relieved by intra-articular administration of mutant mouse TCF12 lentiviral vector (shTCF12). Furthermore, TCF12 upregulated the expression of CXCR4, leading to exacerbation of experimental OA partially through activation of TGF-ß signaling in chondrocytes. TCF12 expression was upregulated in chondrocytes treated with IL-1ß and osteoarthritic chondrocytes. Our findings established an essential role of TCF12 in chondrocyte senescence and cartilage extracellular matrix degradation during OA, and identified intra-articular injection of TCF12 as a potential therapeutic strategy for OA prevention and treatment.

Specific Neural Mechanisms of Self-Cognition and the Application of Brainprint Recognition.

The important identity attribute of self-information presents unique cognitive processing advantages in psychological experiments and has become a research hotspot in psychology and brain science. The unique processing mode of own information has been widely verified in visual and auditory experiments, which is a unique neural processing method for own name, face, voice and other information. In the study of individual behavior, the behavioral uniqueness of self-information is reflected in the faster response of the human brain to self-information, the higher attention to self-information, and the stronger memory level of self-reference. Brain imaging studies have also presented the uniqueness of self-cognition in the brain. EEG studies have shown that self-information induces significant P300 components. fMRI and PET results show that the differences in self and non-self working patterns were located in the frontal and parietal lobes. In addition, this paper combines the self-uniqueness theory and brain-print recognition technology to explore the application of self-information in experimental design, channel combination strategy and identity feature selection of brainprints.

Extremely negative emotion interferes with cognition: Evidence from ERPs and time-varying brain network.

In recent years, the relationship between emotion and cognition was a hot topic. However, it remains unclear which specific emotions can significantly interfere with cognition and how they do so. In this study, we designed a novel Affective Stroop experiment paradigm to investigate these issues. The extremely negative (EN), moderately negative (MN), moderately positive (MP), extremely positive (EP) and neutral pictures were displayed before Stroop tasks. The behavioral results revealed that EN emotion significantly interfered with cognitive performance compared to other types of emotions, with a significant increase in reaction time under the EN emotion condition (P < 0.05). Furthermore, the dynamic brain mechanisms were analyzed from both Event-Related Potential (ERP) and time-varying brain network perspectives. Results showed that EN emotion evoked larger N2, P3, and LPP amplitudes in the frontal, parietal, and occipital brain regions. In contrast, the Stroop task under EN condition led to smaller N2, P3, and LPP amplitudes compared to neutral condition. This indicates that EN emotion was prioritized and consumed more cognitive resources relative to neutral emotion. During the P3 and LPP stages, we observed enhanced bottom-up connections between the parietal and frontal regions while the processing of EN emotion. Additionally, there were stronger top-down cognitive control connections from the frontal to the occipital regions while processing the Stroop task under EN condition. These findings consistently suggest that EN emotion interferes with cognition by consuming more cognitive resources, and the brain needs to enhance cognitive control to support Stroop task execution.

SAST-GCN: Segmentation Adaptive Spatial Temporal-Graph Convolutional Network for P3-Based Video Target Detection.

Detecting video-induced P3 is crucial to building the video target detection system based on the brain-computer interface. However, studies have shown that the brain response patterns corresponding to video-induced P3 are dynamic and determined by the interaction of multiple brain regions. This paper proposes a segmentation adaptive spatial-temporal graph convolutional network (SAST-GCN) for P3-based video target detection. To make full use of the dynamic characteristics of the P3 signal data, the data is segmented according to the processing stages of the video-induced P3, and the brain network connections are constructed correspondingly. Then, the spatial-temporal feature of EEG data is extracted by adaptive spatial-temporal graph convolution to discriminate the target and non-target in the video. Especially, a style-based recalibration module is added to select feature maps with higher contributions and increase the feature extraction ability of the network. The experimental results demonstrate the superiority of our proposed model over the baseline methods. Also, the ablation experiments indicate that the segmentation of data to construct the brain connection can effectively improve the recognition performance by reflecting the dynamic connection relationship between EEG channels more accurately.

ERP-WGAN: A data augmentation method for EEG single-trial detection.

Brain computer interaction based on EEG presents great potential and becomes the research hotspots. However, the insufficient scale of EEG database limits the BCI system performance, especially the positive and negative sample imbalance caused by oddball paradigm. To alleviate the bottleneck problem of scarce EEG sample, we propose a data augmentation method based on generative adversarial network to improve the performance of EEG signal classification. Taking the characteristics of EEG into account in wasserstein generative adversarial networks (WGAN), the problems of model collapse and poor quality of artificial data were solved by using resting noise, smoothing and random amplitude. The quality of artificial data was comprehensively evaluated from verisimilitude, diversity and accuracy. Compared with the three artificial data methods and two data sampling methods, the proposed ERP-WGAN framework significantly improve the performance of both subject and general classifiers, especially the accuracy of general classifiers trained by less than 5 dimensional features is improved by 20-25%. Moreover, we evaluate the training sets performance with different mixing ratios of artificial and real samples. ERP-WGAN can reduced at least 73% of the real subject data and acquisition cost, which greatly saves the test cycle and research cost.

EEG Identity Authentication in Multi-Domain Features: A Multi-Scale 3D-CNN Approach.

Electroencephalogram (EEG) authentication has become a research hotspot in the field of information security due to its advantages of living, internal, and anti-stress. However, the performance of identity authentication system is limited by the inherent attributes of EEG, such as low SNR, low stability, and strong randomness. Researchers generally believe that the in-depth fusion of features can improve the performance of identity authentication and have explored among various feature domains. This experiment invited 70 subjects to participate in the EEG identity authentication task, and the experimental materials were visual stimuli of the self and non-self-names. This paper proposes an innovative EEG authentication framework, including efficient three-dimensional representation of EEG signals, multi-scale convolution structure, and the combination of multiple authentication strategies. In this work, individual EEG signals are converted into spatial-temporal-frequency domain three-dimensional forms to provide multi-angle mixed feature representation. Then, the individual identity features are extracted by the various convolution kernel of multi-scale vision, and the strategy of combining multiple convolution kernels is explored. The results show that the small-size and long-shape convolution kernel is suitable for ERP tasks, which can obtain better convergence and accuracy. The experimental results show that the classification performance of the proposed framework is excellent, and the multi-scale convolution method is effective to extract high-quality identity characteristics across feature domains. The results show that the branch number matches the EEG component number can obtain the excellent cost performance. In addition, this paper explores the network training performance for multi-scale module combination strategy and provides reference for deep network construction strategy of EEG signal processing.

Linking Multi-Layer Dynamical GCN With Style-Based Recalibration CNN for EEG-Based Emotion Recognition.

Electroencephalography (EEG)-based emotion computing has become one of the research hotspots of human-computer interaction (HCI). However, it is difficult to effectively learn the interactions between brain regions in emotional states by using traditional convolutional neural networks because there is information transmission between neurons, which constitutes the brain network structure. In this paper, we proposed a novel model combining graph convolutional network and convolutional neural network, namely MDGCN-SRCNN, aiming to fully extract features of channel connectivity in different receptive fields and deep layer abstract features to distinguish different emotions. Particularly, we add style-based recalibration module to CNN to extract deep layer features, which can better select features that are highly related to emotion. We conducted two individual experiments on SEED data set and SEED-IV data set, respectively, and the experiments proved the effectiveness of MDGCN-SRCNN model. The recognition accuracy on SEED and SEED-IV is 95.08 and 85.52%, respectively. Our model has better performance than other state-of-art methods. In addition, by visualizing the distribution of different layers features, we prove that the combination of shallow layer and deep layer features can effectively improve the recognition performance. Finally, we verified the important brain regions and the connection relationships between channels for emotion generation by analyzing the connection weights between channels after model learning.