Search Results (9,736)

Objectives: Women with Premenstrual Syndrome (PMS) tend to exhibit an excessive attention bias toward negative stimuli during the luteal phase. This study intends to investigate the effect of regular yoga on attention bias of women with PMS during the luteal phase and explore the mechanisms underlying such changes. Methods: Sixty-four women with PMS were recruited, coded and randomly assigned to either a 12-week yoga group (n = 32) or a control group (n = 32). The dot-probe task was used to assess attention bias at baseline and 12 weeks later. Data analysis was performed using SPSS 27.0 software, with analytical methods including descriptive statistics, repeated-measures analysis of variance (RM-ANOVA), simple effect analysis, cluster-based permutation test and Pearson correlation analysis. The Holm–Bonferroni method was used to correct for multiple comparison errors. Results: RM-ANOVA revealed significant time × group interaction effects for attention orientation, attention disengagement, P1 component, and P3 component. Simple effect analysis indicated that, compared with the control group, the yoga group exhibited significant modulations in attention orientation (t = −7.33, p < 0.001), P1 (t = 8.94, p < 0.001), attention disengagement (t = 6.89, p < 0.001), and P3 (t = 4.42, p = 0.002) after 12 weeks of intervention. Cluster-based permutation tests demonstrated that the yoga group showed significant reductions in P1 and P3 amplitudes after 12 weeks. Pearson correlation analysis indicated that attention orientation was significantly negatively correlated with P1 amplitude, while attention disengagement was significantly positively correlated with P3 amplitude. Conclusion: Regular yoga can regulate the behavioral indicators and electroencephalographic (EEG) indicators related to attention bias and exerts a positive effect on modulating attention bias toward negative stimuli in women with PMS during the luteal phase. Full article

Parkinson’s Disease (PD) is a neurodegenerative disorder in which gait abnormalities serve as key indicators of motor impairment and disease progression. Although wearable sensor-based gait analysis has advanced, existing methods still face challenges in modeling multi-sensor spatial relationships, extracting adaptive multi-scale temporal features, and effectively integrating time–frequency information. To address these issues, this paper proposes a multi-sensor gait neural network that integrates biomechanical priors with time–frequency collaborative learning for the automatic assessment of PD gait severity. The framework consists of three core modules: (1) BGS-GAT (Biomechanics-Guided Graph Attention Network), which constructs a sensor graph based on plantar anatomy and explicitly models inter-regional force dependencies via graph attention; (2) AMS-Inception1D (Adaptive Multi-Scale Inception-1D), which employs dilated convolutions and channel attention to extract multi-scale temporal features adaptively; and (3) TF-Branch (Time–Frequency Branch), which applies Real-valued Fast Fourier Transform (RFFT) and frequency-domain convolution to capture rhythmic and high-frequency components, enabling complementary time–frequency representation. Experiments on the PhysioNet multi-channel foot pressure dataset demonstrate that the proposed model achieves 0.930 in accuracy and 0.925 in F1-score for four-class severity classification, outperforming state-of-the-art deep learning models. Full article

Lower limb exoskeleton systems require accurate recognition of the wearer’s movement intentions prior to action execution in order to achieve natural and smooth human–machine interaction. Surface electromyography (sEMG) signals can reflect neural activation of muscles before movement onset, making them a key physiological source for movement intention recognition. To improve sEMG-based recognition performance, this study proposes an innovative deep learning framework, IFIANet. First, a CNN–TCN-based spatiotemporal feature learning network is constructed, which efficiently models and represents multi-scale temporal–frequency features while effectively reducing model parameter complexity. Second, an IFIA (Frequency-Informed Integration Attention) module is designed to incorporate global frequency information, compensating for frequency components potentially lost during time–frequency transformations, thereby enhancing the discriminability and robustness of temporal–frequency features. Extensive ablation and comparative experiments on the publicly available MyPredict1 dataset demonstrate that the proposed framework maintains stable performance across different prediction times and achieves over 82% average recognition accuracy in within-experiments involving nine participants. The results indicate that IFIANet effectively fuses local temporal–frequency features with global frequency priors, providing an efficient and reliable approach for sEMG-based movement intention recognition and intelligent control of exoskeleton systems. Full article

Background/Objective: Neuroimmune and metabolic dysregulation have been increasingly implicated in the cognitive heterogeneity of autism spectrum disorder (ASD). However, it remains unclear whether anti-inflammatory diets engage distinct biological and cognitive pathways in autistic and neurotypical children. This study examined whether a 12-week anti-inflammatory dietary protocol produces group-specific neuroimmune–metabolic signatures and cognitive responses in autistic children, neurotypical children receiving the same diet, and untreated neurotypical controls. Methods: Twenty-two children (11 with ASD, six a on neurotypical diet [NT-diet], and five neurotypical controls [NT-control]) completed pre–post assessments of plasma IFN-γ, CXCL1, RANTES (CCL5), trimethylamine-N-oxide (TMAO), and an extensive ENI-2/WISC-IV neuropsychological battery. Linear mixed-effects models were used to test the Time × Group effects on biomarkers and cognitive domains, adjusting for age, sex, and baseline TMAO. Bayesian estimation quantified individual changes (posterior means, 95% credible intervals, and posterior probabilities). Immune–cognitive coupling was explored using Δ–Δ correlation matrices, network metrics (node strength, degree centrality), exploratory mediation models, and responder (≥0.5 SD domain improvement) versus non-responder analyses. Results: In ASD, the diet induced robust reductions in IFN-γ, RANTES, CXCL1, and TMAO, with decisive Bayesian evidence for IFN-γ and RANTES suppression (posterior P(δ < 0) > 0.99). These shifts were selectively associated with gains in verbal learning, semantic fluency, verbal reasoning, attention, and visuoconstructive abilities, whereas working memory and executive flexibility changes were heterogeneous, revealing executive vulnerability in individuals with smaller TMAO reductions. NT-diet children showed modest but consistent improvements in visuospatial processing, attention, and processing speed, with minimal biomarker changes; NT controls remained biologically and cognitively stable. Network analyses in ASD revealed a dense chemokine-anchored architecture with CXCL1 and RANTES as central hubs linking biomarker reductions to improvements in fluency, memory, attention, and executive flexibility. ΔTMAO predicted changes in executive flexibility only in ASD (explaining >50% of the variance), functioning as a metabolic node of executive susceptibility. Responders displayed larger coordinated decreases in all biomarkers and broader cognitive gains compared to non-responders. Conclusions: A structured anti-inflammatory diet elicits an ASD-specific, coordinated neuroimmune–metabolic response in which suppression of CXCL1 and RANTES and modulation of TMAO are tightly coupled with selective improvements in verbal, attentional, and executive domains. Neurotypical children exhibit modest metabolism-linked cognitive benefits and minimal immune modulation. These findings support a precision-nutrition framework in ASD, emphasizing baseline immunometabolic profiling and network-level biomarkers (CXCL1, RANTES, TMAO) to stratify responders and design combinatorial interventions targeting neuroimmune–metabolic pathways. Full article

For a humanoid robot, it is difficult to predict a motion trajectory through end-to-end imitation learning when performing complex operations and multi-step processes, leading to jittering in the robot arm. To alleviate this problem and reduce the computational complexity of the self-attention module in Vision-Language-Action (VLA) operations, we proposed a memory-gated filtering attention model that improved the multi-head self-attention mechanism. Then, we designed a cross-modal alignment perception during training, combined with a few-shot data-collection strategy for key steps. The experimental results showed that the proposed scheme significantly improved the task success rate and alleviated the robot arm jitter problem, while reducing video memory usage by 72% and improving training speed from 1.35 s to 0.129 s per batch. This maintained higher action accuracy and robustness in the humanoid robot. Full article

Aiming at the problems of small-sized ships (such as small patrol boats) in complex open-sea backgrounds, including small sizes, insufficient feature information, and high missed detection rates, this paper proposes a maritime small target image detection algorithm based on the improved YOLOv11n. Firstly, the BIE module is introduced into the neck feature fusion stage of YOLOv11n. Utilizing its dual-branch information interaction design, independent branches for key features of maritime small targets in infrared and visible light images are constructed, enabling the progressive fusion of infrared and visible light target features. Secondly, RepViTBlock is incorporated into the backbone network and combined with the C3k2 module of YOLOv11n to form C3k2-RepViTBlock. Through the lightweight attention mechanism and multi-branch convolution structure, this addresses the insufficient capture of tiny target features by the C3k2 module and enhances the model’s ability to extract local features of maritime small targets. Finally, the ConvAttn module is embedded at the end of the backbone network. With its dynamic small-kernel convolution, it adaptively extracts the contour features of small targets, maintaining the overall model’s light weight while reducing the missed detection rate for maritime small targets. Experiments on a collected infrared and visible light ship image dataset (IVships) and a public dataset (SeaShips) show that, on the basis of increasing only a small number of parameters, the improved algorithm increases the mAP@0.5 by 1.9% and 1.7%, respectively, and the average precision by 2.2% and 2.4%, respectively, compared with the original model, which significantly improves the model’s small target detection capabilities. Full article

Facial expression recognition (FER) is crucial for affective computing and human–computer interaction; however, it is still difficult to achieve under various conditions in the real world, such as lighting, occlusion, and pose. This work presents a lightweight hybrid network, SE-Hybrid + Face-ViT, which merges convolutional and transformer architectures through multi-level feature fusion and adaptive channel attention. The network includes a convolutional stream to capture the fine-grained texture of the image and a retrained Face-ViT branch to provide the high-level semantic context. Squeeze-and-Excitation (SE) modules adjust the channel responses at different levels, thus allowing the network to focus on the emotion-salient cues and suppress the redundant features. The proposed architecture, trained and tested on the large-scale AffectNet benchmark, achieved 70.45% accuracy and 68.11% macro-F1, thereby outperforming the latest state-of-the-art models such as TBEM-Transformer, FT-CSAT, and HFE-Net by around 2–3%. Grad-CAM-based visualization of the model confirmed accurate attention to the most significant facial areas, resulting in better recognition of subtle expressions such as fear and contempt. The findings indicate that SE-Hybrid + Face-ViT is a computationally efficient yet highly discriminative FER strategy that successfully addresses the issue of how to preserve details while globally reasoning with contextual information locally. Full article

Transcranial focused ultrasound stimulation (tFUS), an emerging non-invasive neuromodulation technique, has garnered growing attention owing to its high spatial resolution and precise targeting capability for deep brain structures. A body of evidence demonstrates that tFUS can effectively modulate neural activity in specific brain regions, inducing excitatory or inhibitory effects, and it is an important means to reshape neural functions. Ultrasound parameters are crucial in determining the transcranial ultrasound modulation effects. However, there is still controversy over which parameters can regulate neural excitability or inhibition, and there are significant differences in the parameters used in previous studies, which have limited the clinical application of transcranial ultrasound to some extent. Therefore, a systematic clarification of parameter–effect relationships is urgently needed to enable qualitative and quantitative understanding of ultrasound-induced neuromodulation, which is essential for achieving reliable and reproducible outcomes. This paper intends to review the effects of different tFUS parameters and their combinations on the excitability and inhibition of brain neural activities as well as the possible mechanisms. By integrating recent findings from both animal models and human clinical studies, we also discuss critical safety issues related to tFUS, aiming to provide a theoretical basis for future transcranial focused ultrasound modulation treatments for various neurological diseases such as stroke, Parkinson’s disease, dementia, epilepsy, pain disorders, and disorders of consciousness while providing reference value for selecting tFUS treatment regimens. Full article

The inspection of power transmission lines using unmanned aerial vehicles primarily relies on object detection. However, the continuous emergence of new obstacle types necessitates frequent updates to detection models, leading to substantial retraining costs. To address this challenge, we propose a novel framework named IViT, which integrates incremental learning with a hybrid CNN-Transformer architecture for improved identification. We combined knowledge distillation with the elastic response selection distillation strategy to enhance detection performance for old classes and strengthen knowledge retention through star convolutional residual blocks constructed via element-wise multiplication. We designed a separable convolution aggregation block that integrates PConv with an attention mechanism, effectively merging global and local information to improve detection accuracy. Finally, we unified the two modules into a hybrid block. In the static detection task, IViT achieves a mAP of 55.3%, a mAP₅₀ of 83.6%, and a mAP₇₅ of 61.0%. For the incremental detection task, it attains a mAP of 57.8%, a mAP₅₀ of 79.7%, and a mAP₇₅ of 62.3%. Extensive experiments on the transmission line corridor external damage dataset and the INSPLAD dataset demonstrate that IViT exhibits outstanding detection performance compared to mainstream static object detection models and incremental object detection models. Full article

This study was conducted to address the demand for interpretable intelligent recognition of fruit tree diseases in smart horticultural environments. A KAD-Former framework integrating an agricultural knowledge graph with a visual Transformer was proposed and systematically validated through extensive cross-regional, multi-variety, and multi-disease experiments. The primary objective of this work was to overcome the limitations of conventional deep models, including insufficient interpretability, unstable recognition of weak disease features, and poor cross-regional generalization. In the experimental evaluation, the model achieved significant advantages across multiple representative tasks: in the overall performance comparison, KAD-Former reached an accuracy of $0.946$, an F1-score of $0.933$, and a mAP of $0.938$, outperforming classical models such as ResNet50, EfficientNet, and Swin-T. In the cross-regional generalization assessment, a DGS of $0.933$ was obtained, notably surpassing competing models. In terms of explainability consistency, a Consistency@5 score of $0.826$ indicated strong alignment between the model’s attention regions and expert annotations. The ablation experiments further demonstrated that the three core modules—AKG (agricultural knowledge graph), SAM (semantic alignment module), and KGA (knowledge-guided attention)—each contributed substantially to final performance, with the complete model exhibiting the best results. These findings collectively demonstrate the comprehensive advantages of KAD-Former in disease classification, symptom localization, model interpretability, and cross-domain transfer. The proposed method not only achieved state-of-the-art performance in pure visual tasks but also advanced knowledge-enhanced and interpretable reasoning by emulating the diagnostic logic employed by agricultural experts in real orchard scenarios. Through the integration of the agricultural knowledge graph, semantic alignment, and knowledge-guided attention, the model maintained stable performance under challenging conditions such as complex illumination, background noise, and weak lesion features, while exhibiting strong robustness in cross-region and cross-variety transfer tests. Furthermore, the experimental results indicated that the approach enhanced fine-grained recognition capabilities for various fruit tree diseases, including apple ring rot, brown spot, powdery mildew, and downy mildew. Full article

Sixth-generation (6G) wireless systems aim to integrate terrestrial, aerial, and satellite networks to support large-scale remote sensing and service delivery. In such non-terrestrial networks (NTNs), channels change quickly and the multi-tier architecture is heterogeneous, which makes real-time channel state acquisition and cooperative resource scheduling difficult. This paper proposes an FMA-MADDPG framework that combines a channel prediction module with a constraint-based multi-agent deep deterministic policy gradient scheme. The Fusion of Mamba and Attention (FMA) predictor uses a Mamba state-space backbone and a multi-head self-attention block to learn both long-term channel evolution and short-term fluctuations, and forecasts future CSI. The predicted channel information is added to the agents’ observations so that scheduling decisions can take expected channel variations into account. A constraint-based reward is also designed, with explicit performance thresholds and anti-idle penalties, to encourage fairness, avoid free-riding, and promote cooperation among heterogeneous agents. In a representative NTN uplink scenario, the proposed method achieves higher total reward, efficiency, load balance, and cooperation than several DRL baselines, with relative gains around 10–20% on key metrics. These results indicate that prediction-aware cooperative reinforcement learning is a useful approach for resource optimization in future 6G NTN systems. Full article

As a critical indicator of polar ecosystem dynamics, sea ice monitoring plays a pivotal role in climate change. However, as global warming accelerates the melting of sea ice, the complexity in the Arctic poses growing challenges for achieving high-precision sea ice classification. To address this issue, this study begins with the creation of a multi-source sea ice dataset based on GaoFen-3 fully polarimetric SAR data and Landsat optical imagery. In addition, the study proposes a Global–Local enhanced Deformable Convolution Network (GLDCN), which effectively captures long-range semantic dependencies and fine-grained local features of sea ice. To further enhance feature integration, an Adaptive Channel Attention Module (ACAM) is designed to achieve adaptive weighted fusion of heterogeneous SAR and optical features, substantially improving the model’s discriminative ability in complex conditions. Experimental results show that the proposed method outperforms several mainstream models on multiple evaluation metrics. The multi-source data fusion strategy significantly reduces misclassification among confusable categories, validating the importance of multimodal fusion in sea ice classification. Full article

In modern high-demand applications, such as real-time video communication, cloud gaming, and high-definition live streaming, achieving both superior transmission speed and high visual fidelity is paramount. However, unstable networks and packet loss remain major bottlenecks, making accurate and low-latency video error concealment a critical challenge. Traditional error control strategies, such as Forward Error Correction (FEC) and Automatic Repeat Request (ARQ), often introduce excessive latency or bandwidth overhead. Meanwhile, receiver-side concealment methods struggle under high motion or significant packet loss, motivating the exploration of predictive models. SimVP-v2, with its efficient convolutional architecture and Gated Spatiotemporal Attention (GSTA) mechanism, provides a strong baseline by reducing complexity and achieving competitive prediction performance. Despite its merits, SimVP-v2’s reliance on 2D convolutions for implicit temporal aggregation limits its capacity to capture complex motion trajectories and long-term dependencies. This often results in artifacts such as motion blur, detail loss, and accumulated errors. Furthermore, its single-modality design ignores the complementary contextual cues embedded in the audio stream. To overcome these issues, we propose A3DSimVP (Audio- and 3D-Enhanced SimVP-v2), which integrates explicit spatio-temporal modeling with multimodal feature fusion. Architecturally, we replace the 2D depthwise separable convolutions within the GSTA module with their 3D counterparts, introducing a redesigned GSTA-3D module that significantly improves motion coherence across frames. Additionally, an efficient audio–visual fusion strategy supplements visual features with contextual audio guidance, thereby enhancing the model’s robustness and perceptual realism. We validate the effectiveness of A3DSimVP’s improvements through extensive experiments on the KTH dataset. Our model achieves a PSNR of 27.35 dB, surpassing the 27.04 of the SimVP-v2 baseline. Concurrently, our improved A3DSimVP model reduces the loss metrics on the KTH dataset, achieving an MSE of 43.82 and an MAE of 385.73, both lower than the baseline. Crucially, our LPIPS metric is substantially lowered to 0.22. These data tangibly confirm that A3DSimVP significantly enhances both structural fidelity and perceptual quality while maintaining high predictive accuracy. Notably, A3DSimVP attains faster inference speeds than the baseline with only a marginal increase in computational overhead. These results establish A3DSimVP as an efficient and robust solution for latency-critical video applications. Full article

Bearings are critical components in industrial machinery, and their failures can lead to equipment downtime and significant safety hazards. Traditional fault diagnosis methods rely on manually crafted features and classical classifiers, often suffering from poor robustness, weak generalization under noisy or small-sample conditions, and limited suitability for lightweight deployment. This study proposes a Lightweight Multi-Scale Multi-Dimensional Self-Attention Transformer (LiMS-MFormer)—an end-to-end lightweight fault diagnosis framework integrating multi-scale feature extraction and multi-dimensional attention. The model integrates lightweight multi-scale convolutional feature extraction, hierarchical feature fusion, and a multi-dimensional self-attention mechanism to balance feature expressiveness with computational efficiency. Specifically, the front end employs Ghost convolution and enhanced residual structures for efficient multi-scale feature extraction. The middle layers perform cross-scale concatenation and fusion to enrich contextual representations. The back end introduces a lightweight temporal-channel-spatial attention module for global modeling and focuses on key patterns. Experiments on the Paderborn University (PU) dataset and the University of Ottawa bearing vibration dataset (Ottawa dataset) show that LiMS-MFormer achieves an accuracy of 96.68% on the small-sample PU dataset while maintaining minimal parameters (0.07 M) and low computational cost (13.55 M FLOPs). Moreover, under complex noisy conditions, the proposed model demonstrates strong fault diagnosis capability. On the University of Ottawa dataset, LiMS-MFormer consistently outperforms several state-of-the-art lightweight models, exhibiting superior accuracy, robustness, and generalization in challenging diagnostic tasks. Full article

Object detection in Unmanned Aerial Vehicle (UAV) imagery has gained significant traction in applications such as railway inspection and waste management. While emerging end-to-end detectors like DEIM show promise, they often struggle with weak feature responses and spatial misalignment in aerial scenarios. To address these issues, this paper proposes SCA-DEIM, a context-aware real-time detection framework. Specifically, we introduce the Adaptive Spatial and Channel Synergistic Attention (ASCSA) module, which refines existing attention paradigms by transitioning from a static gating mechanism to an active signal amplifier. Unlike traditional designs that impose rigid bounds on feature responses, this improved architecture enhances feature extraction by dynamically boosting the saliency of faint small-target signals amidst complex backgrounds. Furthermore, drawing inspiration from infrared small object detection, we propose the Cross-Stage Partial Shifted Pinwheel Mixed Convolution (CSP-SPMConv). By synergizing asymmetric padding with a spatial shift mechanism, this module effectively aligns receptive fields and enforces cross-channel interaction, thereby resolving feature misalignment and scale fusion issues. Comprehensive experiments on the VisDrone2019 dataset demonstrate that, compared with the baseline model, SCA-DEIM achieves improvements of 1.8% in Average Precision (AP), 2.3% in AP for small objects ($A{P}_s$), and 2.0% in AP for large objects ($A{P}_l$), while maintaining a competitive inference speed. Notably, visualization results under different illumination conditions demonstrate the strong robustness of the model. In addition, further validation on both the UAVVaste and UAVDT datasets confirms that the proposed method effectively enhances the detection performance for small objects. Full article