Search Results (197)

The decoding of motor imagery (MI) electroencephalogram (EEG) signals is crucial for motor control and rehabilitation. However, as feature extraction is the core component of the decoding process, traditional methods, often limited to single-feature domains or shallow time-frequency fusion, struggle to comprehensively capture the spatio-temporal-frequency characteristics of the signals, thereby limiting decoding accuracy. To address these limitations, this paper proposes a dual-branch neural network architecture with multi-domain feature fusion, the dual-branch spatio-temporal-frequency fusion convolutional network with Transformer (DB-STFFCNet). The DB-STFFCNet model consists of three modules: the spatiotemporal feature extraction module (STFE), the frequency feature extraction module (FFE), and the feature fusion and classification module. The STFE module employs a lightweight multi-dimensional attention network combined with a temporal Transformer encoder, capable of simultaneously modeling local fine-grained features and global spatiotemporal dependencies, effectively integrating spatiotemporal information and enhancing feature representation. The FFE module constructs a hierarchical feature refinement structure by leveraging the fast Fourier transform (FFT) and multi-scale frequency convolutions, while a frequency-domain Transformer encoder captures the global dependencies among frequency domain features, thus improving the model’s ability to represent key frequency information. Finally, the fusion module effectively consolidates the spatiotemporal and frequency features to achieve accurate classification. To evaluate the feasibility of the proposed method, experiments were conducted on the BCI Competition IV-2a and IV-2b public datasets, achieving accuracies of 83.13% and 89.54%, respectively, outperforming existing methods. This study provides a novel solution for joint time-frequency representation learning in EEG analysis. Full article

Respiratory diseases present significant global health challenges. Recent advances in respiratory sound analysis (RSA) have shown great potential for automated disease diagnosis and patient management. The International Conference on Biomedical and Health Informatics 2017 (ICBHI2017) database stands as one of the most authoritative open-access RSA datasets. This review systematically examines 135 technical publications utilizing the database, and a comprehensive and timely summary of RSA methodologies is offered for researchers and practitioners in this field. Specifically, this review covers signal processing techniques including data resampling, augmentation, normalization, and filtering; feature extraction approaches spanning time-domain, frequency-domain, joint time–frequency analysis, and deep feature representation from pre-trained models; and classification methods for adventitious sound (AS) categorization and pathological state (PS) recognition. Current achievements for AS and PS classification are summarized across studies using official and custom data splits. Despite promising technique advancements, several challenges remain unresolved. These include a severe class imbalance in the dataset, limited exploration of advanced data augmentation techniques and foundation models, a lack of model interpretability, and insufficient generalization studies across clinical settings. Future directions involve multi-modal data fusion, the development of standardized processing workflows, interpretable artificial intelligence, and integration with broader clinical data sources to enhance diagnostic performance and clinical applicability. Full article

Azimuth ambiguity caused by spectral aliasing severely degrades the quality of Synthetic Aperture Radar (SAR) images. To suppress azimuth ambiguity while preserving image details as much as possible, this paper proposes an azimuth ambiguity suppression method for SAR based on time-frequency joint analysis. By exploiting the distribution differences of ambiguous signals across different sub-spectra, the method locates azimuth ambiguity in the time domain through multi-sub-spectrum change detection and fusion, followed by ambiguity suppression in the azimuth time-frequency domain. Experimental results demonstrate that the proposed method effectively suppresses azimuth ambiguity while maintaining superior performance in preserving genuine targets. Full article

Background: Rheumatoid arthritis (RA) is a chronic autoimmune disorder that predominantly affects synovial joints, leading to inflammation, pain, and progressive joint damage. Despite therapeutic advancements, the molecular basis of established RA remains poorly defined. Methods: In this study, we conducted an untargeted plasma proteomic analysis using two-dimensional differential gel electrophoresis (2D-DIGE) and matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF-MS) in samples from RA patients and healthy controls in the discovery phase. Results: Significantly (ANOVA, p ≤ 0.05, fold change > 1.5) differentially abundant proteins (DAPs) were identified. Notably, upregulated proteins included mitochondrial dicarboxylate carrier, hemopexin, and 28S ribosomal protein S18c, while CCDC124, osteocalcin, apolipoproteins A-I and A-IV, and haptoglobin were downregulated. Receiver operating characteristic (ROC) analysis identified CCDC124, osteocalcin, and metallothionein-2 with high diagnostic potential (AUC = 0.98). Proteins with the highest selected frequency were quantitatively verified by multiple reaction monitoring (MRM) analysis in the validation cohort. Bioinformatic analysis using Ingenuity Pathway Analysis (IPA) revealed the underlying molecular pathways and key interaction networks involved STAT1, TNF, and CD40. These central nodes were associated with immune regulation, cell-to-cell signaling, and hematological system development. Conclusions: Our combined proteomic and bioinformatic approaches underscore the involvement of dysregulated immune pathways in RA pathogenesis and highlight potential diagnostic biomarkers. The utility of these markers needs to be evaluated in further studies and in a larger cohort of patients. Full article

Ship seismic wave signals represent one of the most critical physical field characteristics of vessels. To achieve the high-precision detection of ship seismic wave field signals in marine environments, an underwater seismic wave signal detection system was designed. The system adopts a three-stage architecture consisting of watertight instrument housing, a communication circuit, and a buoy to realize high-capacity real-time data transmissions. The host computer performs the collaborative optimization of multi-modal hardware architecture and adaptive signal processing algorithms, enabling the detection of ship targets in oceanic environments. Through verification in a water tank and sea trials, the system successfully measured seismic wave signals. An improved ALE-LOFAR (Adaptive Line Enhancer–Low-Frequency Analysis) joint framework, combined with DEMON (Demodulation of Envelope Modulation) demodulation technology, was proposed to conduct the spectral feature analysis of ship seismic wave signals, yielding the low-frequency signal characteristics of vessels. This scheme provides an important method for the covert monitoring of shallow-sea targets, providing early warnings of illegal fishing and ensuring underwater security. Full article

Autism Spectrum Disorder (ASD) seriously affects social, communication, and behavioral functions, and early accurate diagnosis is crucial to improve the prognosis of patients. Traditional diagnosis methods rely on professional doctors to make subjective diagnosis through scales, the feature extraction of existing machine learning methods is inefficient, and existing deep learning methods have limitations in capturing time-varying features and the joint expression of time–frequency features. To this end, this study proposes a time–frequency synergy network (TFSNet) to improve the accuracy of ASD EEG signal classification. The proposed Dynamic Residual Block (TDRB) was used to enhance time-domain feature extraction; Short-Time Fourier Transform (STFT), convolutional attention mechanism, and transformation technology were combined to capture frequency-domain information; and an adaptive cross-domain attention mechanism (ACDA) was designed to realize efficient fusion of time–frequency features. The experimental results show that the average accuracy of TFSNet on the University of Sheffield (containing 28 ASD patients and 28 healthy controls) and KAU dataset (containing 12 ASD patients and five healthy controls) reaches 98.68%and 97.14%, respectively, yielding significantly better results than the existing machine learning and deep learning methods. In addition, the analysis of model decisions through interpretability analysis techniques enhances its transparency and reliability. Full article

Accurate grid frequency prediction is essential for maintaining the stability and reliability of power systems. However, the complex dynamic characteristics of grid frequency and the nonlinear correlations among massive time series data make it challenging for traditional time series prediction methods to balance efficiency and accuracy. In this paper, we propose a Dynamic Significance–Correlation Weighting (D-SCW) method, which generates dynamic weight coefficients that evolve over time. This is achieved by constructing a joint screening mechanism of feature time series correlation analysis and statistical significance test, combined with the LightGBM gradient-boosting decision tree (GBDT) framework; accordingly, high-precision prediction of grid frequency time series data is realized. To verify the effectiveness of the D-SCW method, this study conducts comparative experiments on two actual grid operation datasets (including typical scenarios with wind/photovoltaic (PV) installations, accounting for 5–35% of the grid); additionally, the Spearman’s rank correlation coefficient method, mutual information (MI), Lasso regression, and the feature screening method of recursive feature elimination (RFE) are selected as the baseline control; root mean square error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE) are adopted as assessment indicators. The results show that the D-SCW-LightGBM framework reduces the root mean squared error (RMSE) by 5.2% to 10.4% and shortens the dynamic response delay by 52% compared with the benchmark method in high renewable penetration scenarios, confirming its effectiveness in both prediction accuracy and computational efficiency. Full article

This paper introduces a novel approach to hand motion gesture recognition by integrating the Fourier transform with hypergraph convolutional networks (HGCNs). Traditional recognition methods often struggle to capture the complex spatiotemporal dynamics of hand gestures. HGCNs, which are capable of modeling intricate relationships among joints, are enhanced by Fourier transform to analyze gesture features in the frequency domain. A hypergraph is constructed to represent the interdependencies among hand joints, allowing for dynamic adjustments based on joint movements. Hypergraph convolution is applied to update node features, while the Fourier transform facilitates frequency-domain analysis. The T-Module, a multiscale temporal convolution module, aggregates features from multiple frames to capture gesture dynamics across different time scales. Experiments on the dynamic hypergraph (DHG14/28) and shape retrieval contest (SHREC’17) datasets demonstrate the effectiveness of the proposed method, achieving accuracies of 96.4% and 97.6%, respectively, and outperforming traditional gesture recognition algorithms. Ablation studies further validate the contributions of each component in enhancing recognition performance. Full article

Thickness inspection of thermal barrier coatings is crucial to safeguard the reliability of high-temperature components of aero-engines, but traditional destructive inspection methods are difficult to meet the demand for rapid assessment in the field. In this study, a new non-destructive testing method integrating terahertz time-domain spectroscopy and machine learning algorithms is proposed to systematically study the thickness inspection of 8YSZ coatings prepared by two processes, namely atmospheric plasma spraying (APS) and electron beam physical vapor deposition (EB-PVD). By optimizing the preparation process parameters, 620 sets of specimens with thicknesses of 100–400 μm are prepared, and three types of characteristic parameters, namely, time delay Δt, frequency shift Δf, and energy decay η, are extracted by combining wavelet threshold denoising and time-frequency joint analysis. A CNN-RF cascade model is constructed to realize coating process classification, and an attention-LSTM and SVR weighted fusion model is developed for thickness regression prediction. The results show that the multimodal feature fusion reduces the root-mean-square error of thickness prediction to 8.9 μm, which further improves the accuracy over the single feature model. The classification accuracy reaches 96.8%, of which the feature importance of time delay Δt accounts for 62%. The hierarchical modeling strategy reduces the detection mean absolute error from 6.2 μm to 4.1 μm. the method provides a high-precision solution for intelligent quality assessment of thermal barrier coatings, which is of great significance in promoting the progress of intelligent manufacturing technology for high-end equipment. Full article

To effectively utilize the rich spectral information of hyperspectral remote sensing images (HRSIs), the fractional Fourier transform (FRFT) feature of HRSIs is proposed to reflect the time-domain and frequency-domain characteristics of a spectral pixel simultaneously, and an FRFT order selection criterion is also proposed based on maximizing separability. Firstly, FRFT is applied to the spectral pixels, and the amplitude spectrum is taken as the FRFT feature of HRSIs. The FRFT feature is mixed with the pixel spectral to form the presented spectral and fractional Fourier transform mixed feature (SF²MF), which contains time–frequency mixing information and spectral information of pixels. K-nearest neighbor, logistic regression, and random forest classifiers are used to verify the superiority of the proposed feature. A 1-dimensional convolutional neural network (1D-CNN) and a two-branch CNN network (Two-CNN_SF²MF-Spa) are designed to extract the depth SF²MF feature and the SF²MF-spatial joint feature, respectively. Moreover, to compensate for the defect that CNN cannot effectively capture the long-range features of spectral pixels, a long short-term memory (LSTM) network is introduced to be combined with CNN to form a two-branch network C-CLSTM_SF²MF for extracting deeper and more efficient fusion features. A 3D-CNN_SF²MF model is designed, which firstly performs the principal component analysis on the spa-SF²MF cube containing spatial information and then feeds it into the 3-dimensional convolutional neural network 3D-CNN_SF²MF to extract the SF²MF-spatial joint feature effectively. The experimental results of three real HRSIs show that the presented mixed feature SF²MF can effectively improve classification accuracy. Full article

In wireless communication systems, traditional frequency synchronization methods struggle to effectively track carrier frequency in low signal-to-noise ratio (SNR) environments, leading to degraded demodulation performance and severely impacting the stability and reliability of communication systems. To address this challenge, an innovative frequency synchronization framework is introduced, enhancing frequency synchronization accuracy and robustness in low-SNR environments through bit synchronization techniques. Specifically, the approach constructs a “bit synchronization-frequency synchronization” joint correction mechanism, where clock offset information extracted during the bit synchronization process is utilized to estimate frequency offset. This method enables an indirect measurement and compensation of carrier frequency offset, forming a hierarchical error compensation system. Furthermore, to overcome the limited convergence speed of the classical Gardner algorithm under significant phase offset conditions, an improved error feedback structure is proposed, accelerating bit synchronization convergence and reducing timing synchronization errors, thereby enhancing overall system performance. The effectiveness of the proposed method is validated through theoretical analysis and simulation experiments. Simulation results demonstrate that, compared to conventional frequency synchronization schemes, the proposed method achieves higher frequency correction accuracy in low-SNR scenarios, thereby improving the robustness and anti-interference capability of wireless communication systems in complex environments. Full article

Background/Objectives: Pre-conception health behaviors may influence the risk of gestational diabetes mellitus (GDM), but evidence on the joint effects of physical activity (PA) and dietary patterns remains limited. This study investigated the associations between pre-conception PA and GDM risk and explored their interaction with adherence to a Mediterranean diet (MD). Methods: This analysis used data from the BORN2020 cohort, which included pregnant women in Greece (2020–2022). Pre-conception PA was assessed using the International Physical Activity Questionnaire-Short Form (IPAQ-SF), expressed as the metabolic equivalent of task (MET)-min/week and categorized into quartiles. Adherence to the MD was assessed via the Trichopoulou score and then grouped into tertiles. Multivariable logistic regression models were computed, accounting for sociodemographic and clinical covariates, including sedentary time and post-lunch nap frequency. Results: In total, 524 women were included and 13.9% (n = 73) were diagnosed with GDM. Women who developed GDM were significantly older (mean age 34.41 vs. 31.98 years, p < 0.0001), were more likely to be >35 years old (46.6% vs. 26.6%, p < 0.001), had higher pre-pregnancy BMI (median 24.6 vs. 22.7 kg/m², p = 0.014), and were more likely to be obese (23.3% vs. 11.8%, p = 0.012). No significant association was observed between total pre-conception PA and GDM risk. Compared to the lowest PA quartile, women in the medium (aOR = 0.80, 95% CI: 0.45–1.40), high (aOR = 1.12, 95% CI: 0.52–2.39), and very high (aOR = 1.10, 95% CI: 0.50–2.38) PA quartiles showed no significant differences in GDM risk. PA, when modeled as a continuous variable, showed no significant trend (aOR = 0.99, 95% CI: 0.99–1.00; p-trend = 0.61). A joint analysis of PA and MD adherence also yielded no significant associations overall. However, in very small BMI-stratified subgroups, a low level of PA combined with very high MD adherence in normal-weight women was associated with increased GDM risk (aOR = 14.06, 95% CI: 1.55–165.54, p = 0.022), while in obese women, very high levels of PA and medium MD adherence showed a potentially protective effect (aOR = 0.006, 95% CI: 8.43 × 10⁻⁶–0.42, p = 0.048). These subgroup findings require cautious interpretation, due to the limited size of the sample set and wide confidence intervals. Conclusions: In this cohort, pre-conception PA, either alone or in combination with MD adherence, was not a reliable predictor of GDM. While our subgroup signals are hypothesis-generating, they do not yet support changes to clinical risk stratification. Future large-scale and interventional studies should investigate combined lifestyle interventions before conception to clarify the potential synergistic effects on GDM prevention. Full article

This paper proposes a joint time–frequency analysis method that combines Rao detector with dynamic sliding thresholds to enhance the detection performance of electric source axial frequency magnetic field signals. For each signal-to-noise ratio (SNR) point, 1000 Monte Carlo simulations were independently conducted, with SNR ranging from 15 dB to −30 dB. The results show that the proposed method maintains high detection rates even at extremely low SNRs, achieving about 90% detection probability at −13 dB, significantly outperforming traditional energy detectors (with a threshold of 2 dB). Under conditions where the detection probability is ≥90% and the false alarm probability is 10⁻³, the SNR threshold for the Rao detector is reduced by 15 dB compared to energy detectors, greatly improving detection performance. Even at lower SNRs (−30 dB), the Rao detector still maintains a certain detection rate, while the detection rate of energy detectors rapidly drops to zero. Further analysis of the impact of different frequencies (1–5 Hz) and CPA distances (45–80 cm) on performance verifies the algorithm’s robustness and practicality in complex non-Gaussian noise environments. This method provides an effective technical solution for low SNR detection of ship axial frequency magnetic fields and has good potential for practical application. Full article

Despite the proliferation of methods for detecting AC arc faults, current solutions often fall short in balancing cost efficiency, real-time performance, and implementation flexibility. This paper proposes a novel joint detection method based on robust local mean decomposition (RLMD) and pulse density analysis. The method leverages a dual-path analog signal processing framework: the low-frequency component of the current transformer (CT) signal is digitized and decomposed using RLMD to extract statistical feature indicators, while the high-frequency component is fed into a high-speed comparator to generate TTL pulses, from which the pulse density is calculated within a sliding time window. By combining the characteristic quantities derived from RLMD with the temporal pulse density, the proposed scheme achieves accurate and efficient detection of arc faults. Experimental results validate the approach, demonstrating high detection probability, adjustable sensitivity, low power consumption, and cost-effectiveness. These attributes underscore the method’s practical relevance and engineering significance in intelligent fault monitoring systems. Full article

In response to the limitations of experimental methods for detecting oil and gas well tubing leaks, this study developed a full-scale indoor simulation system for oil tubing leakage. The system consists of three components: a wellbore simulation device, a dynamic leakage simulation module, and a multi-parameter monitoring system. The wellbore simulator employs a jacketed structure to replicate real-world conditions, while the leakage module incorporates a precision flow control device to regulate leakage rates. The monitoring system integrates high-sensitivity acoustic sensors and pressure sensors. Through multi-condition experiments, the system simulated complex scenarios, including leakage apertures of 1–5 mm, different leakage positions relative to the annular liquid level, and multiple leakage point combinations. A comprehensive acoustic signal processing framework was established, incorporating time–domain features, frequency–domain characteristics, and time–frequency joint analysis. Experimental results indicate that when the leakage point is above the annular liquid level, the acoustic signals received at the wellhead exhibit high-frequency characteristics typical of gas turbulence. In contrast, leaks below the liquid level produce acoustic waves with distinct low-frequency fluid cavitation signatures, accompanied by noticeable medium-coupled attenuation during propagation. These differential features provide a foundation for accurately identifying leakage zones and confirm the feasibility of using acoustic detection technology to locate concealed leaks below the annular liquid level. The study offers experimental support for improving downhole leakage classification and early warning systems. Full article