Electroencephalographic Signal Data Augmentation Based on Improved Generative Adversarial Network.

EEG signals combined with deep learning play an important role in the study of human-computer interaction. However, the limited dataset makes it challenging to study EEG signals using deep learning methods. Inspired by the GAN network in image generation, this paper presents an improved generative adversarial network model L-C-WGAN-GP to generate artificial EEG data to augment training sets and improve the application of BCI in various fields. The generator consists of a long short-term memory (LSTM) network and the discriminator consists of a convolutional neural network (CNN) which uses the gradient penalty-based Wasserstein distance as the loss function in model training. The model can learn the statistical features of EEG signals and generate EEG data that approximate real samples. In addition, the performance of the compressed sensing reconstruction model can be improved by using augmented datasets. Experiments show that, compared with the existing advanced data amplification techniques, the proposed model produces EEG signals closer to the real EEG signals as measured by RMSE, FD and WTD indicators. In addition, in the compressed reconstruction of EEG signals, adding the new data reduces the loss by about 15% compared with the original data, which greatly improves the reconstruction accuracy of the EEG signals' compressed sensing.

A Data Augmentation Method for Motor Imagery EEG Signals Based on DCGAN-GP Network.

Motor imagery electroencephalography (EEG) signals have garnered attention in brain-computer interface (BCI) research due to their potential in promoting motor rehabilitation and control. However, the limited availability of labeled data poses challenges for training robust classifiers. In this study, we propose a novel data augmentation method utilizing an improved Deep Convolutional Generative Adversarial Network with Gradient Penalty (DCGAN-GP) to address this issue. We transformed raw EEG signals into two-dimensional time-frequency maps and employed a DCGAN-GP network to generate synthetic time-frequency representations resembling real data. Validation experiments were conducted on the BCI IV 2b dataset, comparing the performance of classifiers trained with augmented and unaugmented data. Results demonstrated that classifiers trained with synthetic data exhibit enhanced robustness across multiple subjects and achieve higher classification accuracy. Our findings highlight the effectiveness of utilizing a DCGAN-GP-generated synthetic EEG data to improve classifier performance in distinguishing different motor imagery tasks. Thus, the proposed data augmentation method based on a DCGAN-GP offers a promising avenue for enhancing BCI system performance, overcoming data scarcity challenges, and bolstering classifier robustness, thereby providing substantial support for the broader adoption of BCI technology in real-world applications.

Fast reconstruction of EEG signal compression sensing based on deep learning.

When traditional EEG signals are collected based on the Nyquist theorem, long-time recordings of EEG signals will produce a large amount of data. At the same time, limited bandwidth, end-to-end delay, and memory space will bring great pressure on the effective transmission of data. The birth of compressed sensing alleviates this transmission pressure. However, using an iterative compressed sensing reconstruction algorithm for EEG signal reconstruction faces complex calculation problems and slow data processing speed, limiting the application of compressed sensing in EEG signal rapid monitoring systems. As such, this paper presents a non-iterative and fast algorithm for reconstructing EEG signals using compressed sensing and deep learning techniques. This algorithm uses the improved residual network model, extracts the feature information of the EEG signal by one-dimensional dilated convolution, directly learns the nonlinear mapping relationship between the measured value and the original signal, and can quickly and accurately reconstruct the EEG signal. The method proposed in this paper has been verified by simulation on the open BCI contest dataset. Overall, it is proved that the proposed method has higher reconstruction accuracy and faster reconstruction speed than the traditional CS reconstruction algorithm and the existing deep learning reconstruction algorithm. In addition, it can realize the rapid reconstruction of EEG signals.

EEG Emotion Recognition by Fusion of Multi-Scale Features.

Electroencephalogram (EEG) signals exhibit low amplitude, complex background noise, randomness, and significant inter-individual differences, which pose challenges in extracting sufficient features and can lead to information loss during the mapping process from low-dimensional feature matrices to high-dimensional ones in emotion recognition algorithms. In this paper, we propose a Multi-scale Deformable Convolutional Interacting Attention Network based on Residual Network (MDCNAResnet) for EEG-based emotion recognition. Firstly, we extract differential entropy features from different channels of EEG signals and construct a three-dimensional feature matrix based on the relative positions of electrode channels. Secondly, we utilize deformable convolution (DCN) to extract high-level abstract features by replacing standard convolution with deformable convolution, enhancing the modeling capability of the convolutional neural network for irregular targets. Then, we develop the Bottom-Up Feature Pyramid Network (BU-FPN) to extract multi-scale data features, enabling complementary information from different levels in the neural network, while optimizing the feature extraction process using Efficient Channel Attention (ECANet). Finally, we combine the MDCNAResnet with a Bidirectional Gated Recurrent Unit (BiGRU) to further capture the contextual semantic information of EEG signals. Experimental results on the DEAP dataset demonstrate the effectiveness of our approach, achieving accuracies of 98.63% and 98.89% for Valence and Arousal dimensions, respectively.

The inhibitory effect and mechanism of Resina Draconis on the proliferation of MCF-7 breast cancer cells: a network pharmacology-based analysis.

Resina Draconis (RD) is known as the "holy medicine for promoting blood circulation" and possesses antitumor properties against various types of cancer, including breast cancer (BC); however, the underlying mechanism is not well understood. To explore the potential mechanism of RD against BC using network pharmacology and experimental validation, data on bioactive compounds, potential targets of RD, and related genes of BC were obtained from multiple public databases. Gene Ontology (GO) and KEGG pathway analyses were performed via the DAVID database. Protein interactions were downloaded from the STRING database. The mRNA and protein expression levels and survival analysis of the hub targets were analyzed using the UALCAN, HPA, Kaplan‒Meier mapper, and cBioPortal databases. Subsequently, molecular docking was used to verify the selected key ingredients and hub targets. Finally, the predicted results of network pharmacology methods were verified by cell experiments. In total, 160 active ingredients were obtained, and 148 RD target genes for the treatment of BC were identified. KEGG pathway analysis indicated that RD exerted its therapeutic effects on BC by regulating multiple pathways. Of these, the PI3K-AKT pathway was indicated to play an important role. In addition, RD treatment of BC seemed to involve the regulation of hub targets that were identified based on PPI interaction network analysis. Validation in different databases showed that AKT1, ESR1, HSP90AA1, CASP3, SRC and MDM2 may be involved in the carcinogenesis and progression of BC and that ESR1, IGF1 and HSP90AA1 were correlated with worse overall survival (OS) in BC patients. Molecular docking results showed that 103 active compounds have good binding activity with the hub targets, among which flavonoid compounds were the most important active components. Therefore, the sanguis draconis flavones (SDF) were selected for subsequent cell experiments. The experimental results showed that SDF significantly inhibited the cell cycle and cell proliferation of MCF-7 cells through the PI3K/AKT pathway and induced MCF-7 cell apoptosis. This study has preliminarily reported on the active ingredients, potential targets, and molecular mechanism of RD against BC, and RD was shown to exert its therapeutic effects on BC by regulating the PI3K/AKT pathway and related gene targets. Importantly, our work could provide a theoretical basis for further study of the complex anti-BC mechanism of RD.

Digital-Twin-Assisted Edge-Computing Resource Allocation Based on the Whale Optimization Algorithm.

With the rapid increase of smart Internet of Things (IoT) devices, edge networks generate a large number of computing tasks, which require edge-computing resource devices to complete the calculations. However, unreasonable edge-computing resource allocation suffers from high-power consumption and resource waste. Therefore, when user tasks are offloaded to the edge-computing system, reasonable resource allocation is an important issue. Thus, this paper proposes a digital-twin-(DT)-assisted edge-computing resource-allocation model and establishes a joint-optimization function of power consumption, delay, and unbalanced resource-allocation rate. Then, we develop a solution based on the improved whale optimization scheme. Specifically, we propose an improved whale optimization algorithm and design a greedy initialization strategy to improve the convergence speed for the DT-assisted edge-computing resource-allocation problem. Additionally, we redesign the whale search strategy to improve the allocation results. Several simulation experiments demonstrate that the improved whale optimization algorithm reduces the resource allocation and allocation objective function value, the power consumption, and the average resource allocation imbalance rate by 12.6%, 15.2%, and 15.6%, respectively. Overall, the power consumption with the assistance of the DT is reduced to 89.6% of the power required without DT assistance, thus, improving the efficiency of the edge-computing resource allocation.

Identification of medicinal plants within the Apocynaceae family using ITS2 and psbA-trnH barcodes.

To ensure the safety of medications, it is vital to accurately authenticate species of the Apocynaceae family, which is rich in poisonous medicinal plants. We identified Apocynaceae species by using nuclear internal transcribed spacer 2 (ITS2) and psbA-trnH based on experimental data. The identification ability of ITS2 and psbA-trnH was assessed using specific genetic divergence, BLAST1, and neighbor-joining trees. For DNA barcoding, ITS2 and psbA-trnH regions of 122 plant samples of 31 species from 19 genera in the Apocynaceae family were amplified. The PCR amplification for ITS2 and psbA-trnH sequences was 100%. The sequencing success rates for ITS2 and psbA-trnH sequences were 81% and 61%, respectively. Additional data involved 53 sequences of the ITS2 region and 38 sequences of the psbA-trnH region were downloaded from GenBank. Moreover, the analysis showed that the inter-specific divergence of Apocynaceae species was greater than its intra-specific variations. The results indicated that, using the BLAST1 method, ITS2 showed a high identification efficiency of 97% and 100% of the samples at the species and genus levels, respectively, via BLAST1, and psbA-trnH successfully identified 95% and 100% of the samples at the species and genus levels, respectively. The barcode combination of ITS2/psbA-trnH successfully identified 98% and 100% of samples at the species and genus levels, respectively. Subsequently, the neighbor joining tree method also showed that barcode ITS2 and psbA-trnH could distinguish among the species within the Apocynaceae family. ITS2 is a core barcode and psbA-trnH is a supplementary barcode for identifying species in the Apocynaceae family. These results will help to improve DNA barcoding reference databases for herbal drugs and other herbal raw materials.

Efficacy of Obcordata A from Aspidopterys obcordata on Kidney Stones by Inhibiting NOX4 Expression.

Obcordata A (OA) is a polyoxypregnane glycoside derived from the Dai medicine Aspidopterys obcordata vines. This study aims to investigate the efficacy of OA on renal tubular epithelial cells exposed to calcium oxalate crystals. We incubated renal tubular cells with 28 µg·cm2 calcium oxalate crystals for 24 h with and without OA, GKT137831, phorbol-12-myristate-13-acetate (PMA), and tocopherol. The MTT [3-(4, 5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] assay, microscopic examination, flow cytometry, and immunofluorescence staining revealed that calcium oxalate crystals decreased cell viability and elevated reactive oxygen species (ROS) levels. OA, GKT137831, and tocopherol protected cells and decreased ROS levels. However, OA did not exhibit direct DPPH scavenging ability. In addition, immunoblotting illustrated that OA inhibited the NOX4 (nicotinamide adenine dinucleotide phosphate oxidases 4) expression and downregulated the protein expression in the NOX4/ROS/p38 MAPK (p38 mitogen-activated protein kinase) pathway. The findings suggest that the cytoprotective and antioxidant effects of OA can be blocked by the NOX4 agonist PMA. In conclusion, OA could be used as a NOX4 inhibitor to prevent kidney stones.

Bioinformatics facilitating the use of microarrays to delineate potential miRNA biomarkers in aristolochic acid nephropathy.

Aristolochic acid nephropathy (AAN) is a rapidly progressive acute or chronic tubulointerstitial nephritis (TIN). The present study attempted to explore the molecular mechanisms underlying the miRNA-directed development of AAN. Our differentially expressed analysis identified 11 DE-miRNAs and retrieved the target genes of these DE-miRNAs; then, network analysis and functional analysis further identified 6 DE-miRNAs (has-miR-192, has-miR-194, has-miR-542-3p, has-miR-450a, has-miR-584, has-miR-33a) as phenotypic biomarkers of AAN. Surprisingly, of has-miR-192 has been reported to be associated with the pathogenesis of AAN, and has-miR-194, has-miR-542-3p and has-miR-450a was first-time identified to link to the development of AAN. In addition, the expressional changes of has-miR-584 and has-miR-33a may be associated with the development of AAN as well, which must be further confirmed by the associated experiments. Taken together, our work reveals for the first time the regulatory mechanisms of miRNAs in the development of AAN and this will contribute to miRNA-based diagnosis and treatment of AAN.

Prediction and Characterisation of the System Effects of Aristolochic Acid: A Novel Joint Network Analysis towards Therapeutic and Toxicological Mechanisms.

Aristolochic acid (AA) is the major active component of medicinal plants from the Aristolochiaceae family of flowering plants widely utilized for medicinal purposes. However, the molecular mechanisms of AA systems effects remain poorly understood. Here, we employed a joint network analysis that combines network pharmacology, a protein-protein interaction (PPI) database, biological processes analysis and functional annotation analysis to explore system effects. Firstly, we selected 15 protein targets (14 genes) in the PubChem database as the potential target genes and used PPI knowledge to incorporate these genes into an AA-specific gene network that contains 129 genes. Secondly, we performed biological processes analysis for these AA-related targets using ClueGO, some of new targeted genes were randomly selected and experimentally verified by employing the Quantitative Real-Time PCR assay for targeting the systems effects of AA in HK-2 cells with observed dependency of concentration. Thirdly, the pathway-based functional enrichment analysis was manipulated using WebGestalt to identify the mostly significant pathways associated with AA. At last, we built an AA target pathway network of significant pathways to predict the system effects. Taken together, this joint network analysis revealed that the systematic regulatory effects of AA on multidimensional pathways involving both therapeutic action and toxicity.

Prioritizing disease candidate proteins in cardiomyopathy-specific protein-protein interaction networks based on "guilt by association" analysis.

The cardiomyopathies are a group of heart muscle diseases which can be inherited (familial). Identifying potential disease-related proteins is important to understand mechanisms of cardiomyopathies. Experimental identification of cardiomyophthies is costly and labour-intensive. In contrast, bioinformatics approach has a competitive advantage over experimental method. Based on "guilt by association" analysis, we prioritized candidate proteins involving in human cardiomyopathies. We first built weighted human cardiomyopathy-specific protein-protein interaction networks for three subtypes of cardiomyopathies using the known disease proteins from Online Mendelian Inheritance in Man as seeds. We then developed a method in prioritizing disease candidate proteins to rank candidate proteins in the network based on "guilt by association" analysis. It was found that most candidate proteins with high scores shared disease-related pathways with disease seed proteins. These top ranked candidate proteins were related with the corresponding disease subtypes, and were potential disease-related proteins. Cross-validation and comparison with other methods indicated that our approach could be used for the identification of potentially novel disease proteins, which may provide insights into cardiomyopathy-related mechanisms in a more comprehensive and integrated way.