Search Results (437)

Accurate evaluation of fish feeding intensity is crucial for optimizing aquaculture efficiency and the healthy growth of fish. Previous methods mainly rely on single-modal approaches (e.g., audio or visual). However, the complex underwater environment makes single-modal monitoring methods face significant challenges: visual systems are severely affected by water turbidity, lighting conditions, and fish occlusion, while acoustic systems suffer from background noise. Although existing studies have attempted to combine acoustic and visual information, most adopt simple feature-level fusion strategies, which fail to fully explore the complementary advantages of the two modalities under different environmental conditions and lack dynamic evaluation mechanisms for modal reliability. To address these problems, we propose the Adaptive Cross-modal Attention Fusion Network (ACAF-Net), a cross-modal complementarity learning framework with a two-stage attention fusion mechanism: (1) a cross-modal enhancement stage that enriches individual representations through Low-rank Bilinear Pooling and learnable fusion weights; (2) an adaptive attention fusion stage that dynamically weights acoustic and visual features based on complementarity and environmental reliability. Our framework incorporates dimension alignment strategies and attention mechanisms to capture temporal–spatial complementarity between acoustic feeding signals and visual behavioral patterns. Extensive experiments demonstrate superior performance compared to single-modal and conventional fusion approaches, with 6.4% accuracy improvement. The results validate the effectiveness of exploiting cross-modal complementarity for underwater behavioral analysis and establish a foundation for intelligent aquaculture monitoring systems. Full article

This study proposes a Swin-ReshoUnet architecture with a three-level enhancement mechanism to address inefficiencies in multi-scale feature extraction and gradient degradation in deep networks for high-precision seismic exploration. The encoder uses a hierarchical convolution module to build a multi-scale feature pyramid, enhancing cross-scale geological signal representation. The decoder replaces traditional self-attention with ORCA attention to enable global context modeling with lower computational cost. Skip connections integrate a residual channel attention module, mitigating gradient degradation via dual-pooling feature fusion and activation optimization, forming a full-link optimization from low-level feature enhancement to high-level semantic integration. Simulated and real dataset experiments show that at decimation ratios of 0.1–0.5, the method significantly outperforms SwinUnet, TransUnet, etc., in reconstruction performance. Residual signals and F-K spectra verify high-fidelity reconstruction. Despite increased difficulty with higher sparsity, it maintains optimal performance with notable margins, demonstrating strong robustness. The proposed hierarchical feature enhancement and cross-scale attention strategies offer an efficient seismic profile signal reconstruction solution and show generality for migration to complex visual tasks, advancing geophysics-computer vision interdisciplinary innovation. Full article

Underwater acoustic (UWA) communication supports many critical applications but still faces several physical-layer signal processing challenges. In response, recent advances in machine learning (ML) and deep learning (DL) offer promising solutions to improve signal detection, modulation adaptability, and classification accuracy. These developments highlight the need for a systematic evaluation to compare various ML/DL models and assess their performance across diverse underwater conditions. However, most existing reviews on ML/DL-based UWA communication focus on isolated approaches rather than integrated system-level perspectives, which limits cross-domain insights and reduces their relevance to practical underwater deployments. Consequently, this systematic literature review (SLR) synthesizes 43 studies (2020–2025) on ML and DL approaches for UWA communication, covering channel estimation, adaptive modulation, and modulation recognition across both single- and multi-carrier systems. The findings reveal that models such as convolutional neural networks (CNNs), long short-term memory networks (LSTMs), and generative adversarial networks (GANs) enhance channel estimation performance, achieving error reductions and bit error rate (BER) gains ranging from ${10}^{-3}$ to ${10}^{-6}$. Adaptive modulation techniques incorporating support vector machines (SVMs), CNNs, and reinforcement learning (RL) attain classification accuracies exceeding 98% and throughput improvements of up to 25%. For modulation recognition, architectures like sequence CNNs, residual networks, and hybrid convolutional–recurrent models achieve up to 99.38% accuracy with latency below 10 ms. These performance metrics underscore the viability of ML/DL-based solutions in optimizing physical-layer tasks for real-world UWA deployments. Finally, the SLR identifies key challenges in UWA communication, including high complexity, limited data, fragmented performance metrics, deployment realities, energy constraints and poor scalability. It also outlines future directions like lightweight models, physics-informed learning, advanced RL strategies, intelligent resource allocation, and robust feature fusion to build reliable and intelligent underwater systems. Full article

Ergonomic risk classification during manual lifting tasks is crucial for the prevention of workplace injuries. This study addresses the challenge of classifying lifting task risk levels (low, medium, and high risk, labeled as 0, 1, and 2) using multi-modal time-series data comprising raw facial landmarks and bio-signals (electrocardiography [ECG] and electrodermal activity [EDA]). Classifying such data presents inherent challenges due to multi-source information, temporal dynamics, and class imbalance. To overcome these challenges, this paper proposes a Multi-Adaptive Functional Neural Network (Multi-AdaFNN), a novel method that integrates functional data analysis with deep learning techniques. The proposed model introduces a novel adaptive basis layer composed of micro-networks tailored to each individual time-series feature, enabling end-to-end learning of discriminative temporal patterns directly from raw data. The Multi-AdaFNN approach was evaluated across five distinct dataset configurations: (1) facial landmarks only, (2) bio-signals only, (3) full fusion of all available features, (4) a reduced-dimensionality set of 12 selected facial landmark trajectories, and (5) the same reduced set combined with bio-signals. Performance was rigorously assessed using 100 independent stratified splits (70% training and 30% testing) and optimized via a weighted cross-entropy loss function to manage class imbalance effectively. The results demonstrated that the integrated approach, fusing facial landmarks and bio-signals, achieved the highest classification accuracy and robustness. Furthermore, the adaptive basis functions revealed specific phases within lifting tasks critical for risk prediction. These findings underscore the efficacy and transparency of the Multi-AdaFNN framework for multi-modal ergonomic risk assessment, highlighting its potential for real-time monitoring and proactive injury prevention in industrial environments. Full article

Maternal obesity programs the fetus for increased risk of chronic disease development in early life and adulthood. We hypothesized that maternal nutrient excess leads to fetal inflammation and impairs offspring skeletal muscle mitochondrial biogenesis in non-human primates. At least 12 months before pregnancy, female baboons were fed a normal chow (CTR, 12% energy fat) or a maternal nutrient excess (MNE, 45% energy fat, and ad libitum fructose sodas) diet, with the latter to induce obesity. After 165 days of gestation (0.9 G), offspring baboons were delivered by cesarean section, and the soleus muscle was collected (CTR n = 16, MNE n = 5). At conception, MNE mothers presented increased body fat and weighed more than controls. The soleus muscle of MNE fetuses exhibited increased levels of stress signaling associated with inflammation (TLR4, TNFα, NF-kB p65, and p38), concomitant with reduced expression of key regulators of mitochondrial biogenesis, including PGC1α, both at the protein and transcript levels, as well as downregulation of PPARGC1B, PPARA, PPARB, CREB1, NOS3, SIRT1, SIRT3. Decreased transcript levels of NRF1 were observed alongside diminished mitochondrial DNA copy number, mitochondrial fusion elements (MFN1, MFN2), cytochrome C protein levels, and cytochrome C oxidase subunits I and II transcripts (cox1 and cox2). MNE coupled to MO-induced stress signaling in fetal baboon soleus muscle is associated with impaired mitochondrial biogenesis and lower mitochondrial content, resembling the changes observed in metabolic dysfunctions, such as diabetes. The observed fetal alterations may have important implications for postnatal development and metabolism, potentially increasing the risk of early-onset metabolic disorders and other non-communicable diseases. Full article

This study proposes MSFF-Net, a lightweight deep learning framework for bearing fault diagnosis based on frequency-domain multi-sensor fusion. The vibration and acoustic signals are initially converted into the frequency domain using the fast Fourier transform (FFT), enabling the extraction of temporally invariant spectral features. These features are processed by a compact one-dimensional convolutional neural network, where modality-specific representations are fused at the feature level to capture complementary fault-related information. The proposed method demonstrates robust and superior performance under both full and scarce data conditions, as verified through experiments on a publicly available dataset. Experimental results on a publicly available dataset indicate that the proposed model attains an average accuracy of 99.73%, outperforming state-of-the-art (SOTA) methods in both accuracy and stability. With only about 70.3% of the parameters of the SOTA model, it offers faster inference and reduced computational cost. Ablation studies confirm that multi-sensor fusion improves all classification metrics over single-sensor setups. Under few-shot conditions with 20 samples per class, the model retains 94.69% accuracy, highlighting its strong generalization in data-limited scenarios. The results validate the effectiveness, computational efficiency, and practical applicability of the model for deployment in data-constrained industrial environments. Full article

Sweet taste plays a pivotal role in human dietary behavior and metabolic regulation. With the increasing incidence of metabolic disorders linked to excessive sugar intake, the development and accurate evaluation of new sweeteners have become critical topics in food science and public health. However, the structural diversity of sweeteners and their complex interactions with sweet taste receptors present major challenges for standardized sweetness detection. This review offers a comprehensive and up-to-date overview of sweet taste transmission mechanisms and current detection methods. It outlines the classification and sensory characteristics of both conventional and emerging sweeteners, and explains the multi-level signaling pathway from receptor binding to neural encoding. Key detection techniques, including sensory evaluation, electronic tongues, and biosensors, are systematically compared in terms of their working principles, application scope, and limitations. Special emphasis is placed on advanced biosensing technologies utilizing receptor–ligand interactions and nanomaterials for highly sensitive and specific detection. Furthermore, an intelligent detection framework integrating molecular recognition, multi-source data fusion, and artificial intelligence is proposed. This interdisciplinary approach provides new insights and technical solutions to support precise sweetness evaluation and the future development of healthier food systems. Full article

The quantitative assessment of Parkinson’s disease (PD) is critical for guiding diagnosis, treatment, and rehabilitation. Conventional clinical evaluations—heavily dependent on manual rating scales such as the Unified Parkinson’s Disease Rating Scale (UPDRS)—are time-consuming and prone to inter-rater variability. In this study, we propose a fully automated UPDRS gait-scoring framework. Our method combines (a) surface electromyography (EMG) signals and (b) inertial measurement unit (IMU) data into a single deep learning model. Our end-to-end network comprises three specialized branches—a diagnosis head, an evaluation head, and a balance head—whose outputs are integrated via a customized fusion-detection module to emulate the multidimensional assessments performed by clinicians. We validated our system on 21 PD patients and healthy controls performing a simple walking task while wearing a four-channel EMG array on the lower limbs and 2 shank-mounted IMUs. It achieved a mean classification accuracy of 92.8% across UPDRS levels 0–2. This approach requires minimal subject effort and sensor setup, significantly cutting clinician workload associated with traditional UPDRS evaluations while improving objectivity. The results demonstrate the potential of wearable sensor-driven deep learning methods to deliver rapid, reliable PD gait assessment in both clinical and home settings. Full article

The object of this research is the process of tracking air targets at the initial stage of radar system operation. The problem lies in the lack of a comprehensive approach to tracking air targets in difficult conditions that is able to dynamically adapt filtering parameters, predict signal reliability, and change the processing mode depending on the level of interference. In conditions of signal loss, noise, and unstable measurement reliability, traditional methods do not provide stable and accurate tracking, especially at the initial stages of radar operation. To address this issue, an intelligent method is proposed that integrates a probabilistic graphical evaluation and review technique (GERT) model, a recursive Kalman filter, and a measurement reliability prediction module based on a long short-term memory (LSTM) neural network. The proposed approach allows for the real-time adaptation of filtering parameters, fusion of local and global trajectory estimates, and dynamic switching between tracking modes depending on the environmental conditions. The dynamic weighting algorithm between model estimates ensures a balance between accuracy and robustness. Simulation experiments confirmed the effectiveness of the method: the root mean square error (RMSE) of coordinate estimation was reduced by 25%; the probability of tracking loss decreased by half (from 0.2 to 0.1); and the accuracy of loss prediction exceeded 85%. The novelty of the approach lies in integrating stochastic modeling, machine learning, and classical filtering into a unified adaptive loop. The proposed system can be adapted to various types of radar and easily scaled to multi-sensor architectures. This makes it suitable for practical implementation in both defense and civilian air object detection systems operating under complex conditions. Full article

This study investigates the role of melatonin in alleviating the oxidative stress and apoptosis of TM3 Leydig cells induced by 4-methyl-2,4-bis(4-hydroxyphenyl)pent-1-ene (MBP), the primary active metabolite of Bisphenol A, and clarifies its potential mechanisms involving the SIRT1/PGC-1α pathway. We found that melatonin effectively mitigated MBP-induced cytotoxicity in TM3 cells (p < 0.05). The testosterone levels and steroid hormone synthesis proteins were significantly restored by melatonin. Furthermore, there was a significant reduction in apoptosis after melatonin treatment both in MBP-treated TM3 cells and Bisphenol A-treated testicular interstitial tissues (p < 0.05), along with a significant decrease in the pro-apoptotic markers Bax and cleaved caspase 3, and a significant increase in the anti-apoptotic Bcl-2 level and the Bcl-2/Bax ratio in TM3 cells (p < 0.05). Additionally, the mitochondrial membrane potential improved significantly, ROS and MDA levels were down-regulated, and ATP production was elevated following melatonin treatment in TM3 cells. Mechanistically, melatonin promoted PGC-1α expression and activated the SIRT1 signaling pathway in MBP-treated TM3 cells and Bisphenol A-treated testicular interstitial tissues. This leads to increased expression of NRF2 and its downstream antioxidant genes, mitochondrial respiratory chain complex-related genes, mitochondrial biogenesis genes, and mitochondrial fusion genes while significantly reducing mitochondrial fission genes (p < 0.05). The PGC-1α inhibitor SR-18292 reversed these protective effects, confirming the critical role of this pathway. Conclusively, melatonin exerts a protective effect against MBP-induced oxidative stress and apoptosis in TM3 cells through the SIRT1/PGC-1α pathway, indicating its potential as a therapeutic agent for improving male reproductive health compromised by environmental toxins. Full article

Urban traffic management faces challenges, including inadequate sensing capabilities and insufficient operational status evaluation. The rapid expansion of electronic toll collection (ETC) systems from highways to urban roads provides new opportunities to address these issues. The vast amount of “dormant” ETC data contains rich traffic information that urgently needs to be deeply mined and effectively utilized. This paper reviews the research status, key technologies, and development trends of urban traffic state sensing and analysis technologies based on ETC data. In terms of technological development, ETC systems have evolved from simple toll collection tools to comprehensive traffic management platforms, featuring unique advantages such as accurate vehicle identification, extensive spatiotemporal coverage, and stable data quality. ETC data-based traffic sensing technologies encompass traffic state representation at microscopic, mesoscopic, and macroscopic levels, enabling comprehensive sensing from individual vehicle behavior to overall network operations. The construction of multi-source data fusion frameworks enables effective complementarity between ETC data, floating car data, and video detection data, significantly improving traffic state estimation accuracy. In practical applications, ETC data has demonstrated enormous potential in real-time monitoring and signal control optimization, traffic prediction and artificial intelligence technologies, environmental impact assessment, and other fields. Meanwhile, ETC data-based urban traffic management is transitioning from passive responses to proactive prediction, from single functions to comprehensive services, and from isolated systems to integrated platforms. Looking toward the future, the deep integration of emerging technologies, such as vehicle–road networking, edge computing, and artificial intelligence, with ETC systems will further promote the intelligent, refined, and precise development of urban traffic management. Full article

With the rapid progress of intelligent vehicle technology, the accurate recognition of road surface types and conditions has emerged as a crucial technology for improving the safety and comfort levels in autonomous driving. This paper puts forward a multi-feature fusion approach for road surface identification. Relying on a 24 GHz millimeter-wave radar, statistical features are combined with wavelet transform techniques. This combination enables the efficient classification of diverse road surface types and conditions. Firstly, the discriminability of radar echo signals corresponding to different road surface types is verified via statistical analysis. During this process, six-dimensional statistical features that display remarkable differences are extracted. Subsequently, a novel radar data reconstruction approach is presented. This method involves fitting discrete echo signals into coordinate curves. Then, discrete wavelet transform is utilized to extract both low-frequency and high-frequency features, thereby strengthening the spatio-temporal correlation of the signals. The low-frequency information serves to capture general characteristics, whereas the high-frequency information reflects detailed features. The statistical features and wavelet transform features are fused at the feature level, culminating in the formation of a 56-dimensional feature vector. Four machine learning models, namely the Wide Neural Network (WNN), K-Nearest Neighbors (KNN), Support Vector Machine (SVM), and Kernel methods, are employed as classifiers for both training and testing purposes. Experiments were executed with 8865 samples obtained from a real-vehicle platform. These samples comprehensively represented 12 typical road surface types and conditions. The experimental outcomes clearly indicate that the proposed method is capable of attaining a road surface type identification accuracy as high as 94.2%. As a result, it furnishes an efficient and cost-efficient road perception solution for intelligent driving systems. This research validates the potential application of millimeter-wave radar in intricate road environments and offers both theoretical underpinning and practical support for the advancement of autonomous driving technology. Full article

Autophagy is impaired in Alzheimer’s disease (AD), particularly at the stage of autophagosome–lysosome fusion. Recent studies suggest that the inositol polyphosphate 5-phosphatase OCRL (Lowe oculocerebrorenal syndrome protein) is involved in this fusion process; however, its role in AD pathophysiology remains largely unclear. In this study, we investigated the localization and expression of OCRL in post-mortem AD brains and in a 5XFAD transgenic mouse model. While OCRL RNA levels were not significantly altered, OCRL protein was markedly reduced in the RIPA-soluble fraction and positively correlated with the autophagy marker Beclin1. Immunohistochemical analysis revealed OCRL immunoreactivity in neuronal cytoplasm, granulovacuolar degeneration bodies, and plaque-associated dystrophic neurites in AD brains. Furthermore, OCRL overexpression in a FRET-based tau biosensor cell model significantly reduced the tau-seeding-induced FRET signal. These findings suggest that OCRL dysregulation may contribute to autophagic deficits and the progression of tau pathology in AD. Full article

In real-world scenarios, the rotational speed of bearings is variable. Due to changes in operating conditions, the feature distribution of bearing vibration data becomes inconsistent, which leads to the inability to directly apply the training model built under one operating condition (source domain) to another condition (target domain). Furthermore, the lack of sufficient labeled data in the target domain further complicates fault diagnosis under varying operating conditions. To address this issue, this paper proposes a spatiotemporal feature fusion domain-adaptive network (STFDAN) framework for bearing fault diagnosis under varying operating conditions. The framework constructs a feature extraction and domain adaptation network based on a parallel architecture, designed to capture the complex dynamic characteristics of vibration signals. First, the Fast Fourier Transform (FFT) and Variational Mode Decomposition (VMD) are used to extract the spectral and modal features of the signals, generating a joint representation with multi-level information. Then, a parallel processing mechanism of the Convolutional Neural Network (SECNN) based on the Squeeze-and-Excitation module and the Bidirectional Long Short-Term Memory network (BiLSTM) is employed to dynamically adjust weights, capturing high-dimensional spatiotemporal features. The cross-attention mechanism enables the interaction and fusion of spatial and temporal features, significantly enhancing the complementarity and coupling of the feature representations. Finally, a Multi-Kernel Maximum Mean Discrepancy (MKMMD) is introduced to align the feature distributions between the source and target domains, enabling efficient fault diagnosis under varying bearing conditions. The proposed STFDAN framework is evaluated using bearing datasets from Case Western Reserve University (CWRU), Jiangnan University (JNU), and Southeast University (SEU). Experimental results demonstrate that STFDAN achieves high diagnostic accuracy across different load conditions and effectively solves the bearing fault diagnosis problem under varying operating conditions. Full article

Large pre-trained models (PLMs) have provided tremendous opportunities and potentialities for multimodal fake news detection. However, existing multimodal fake news detection methods never manipulate the token-wise hierarchical semantics of news yielded from PLMs and extremely rely on contrastive learning but ignore the symmetry between text and image in terms of the abstract level. This paper proposes a novel multimodal fake news detection method that helps to balance the understanding between text and image via (1) designing a global-token across-attention mechanism to capture the correlations between global text and tokenwise image representations (or tokenwise text and global image representations) obtained from BERT and ViT; (2) proposing a QK-sharing strategy within cross-attention to enforce model symmetry that reduces information redundancy and accelerates fusion without sacrificing representational power; (3) deploying a semantic augmentation module that systematically extracts token-wise multilayered text semantics from stacked BERT blocks via CNN and Bi-LSTM layers, thereby rebalancing abstract-level disparities by symmetrically enriching shallow and deep textual signals. We also prove the effectiveness of our approach by comparing it with four state-of-the-art baselines. All the comparisons were conducted using three widely adopted multimodal fake news datasets. The results show that our approach outperforms the benchmarks by 0.8% in accuracy and 2.2% in F1-score on average across the three datasets, which demonstrates a symmetric, token-centric fusion of fine-grained semantic fusion, thereby driving more robust fake news detection. Full article