Search Results (497)

Recently, deep learning algorithms have been increasingly applied in construction for activity recognition, particularly for excavators, to automate processes and enhance safety and productivity through continuous monitoring of earthmoving activities. These deep learning algorithms analyze construction videos to classify excavator activities for earthmoving purposes. However, previous studies have solely focused on single-source external videos, which limits the activity recognition capabilities of the deep learning algorithm. This paper introduces a novel multi-modal deep learning-based methodology for recognizing excavator activities, utilizing multi-stream input data. It processes point clouds and RGB images using the two-stream long short-term memory convolutional neural network (CNN-LSTM) method to extract spatiotemporal features, enabling the recognition of excavator activities. A comprehensive dataset comprising 495,000 video frames of synchronized RGB and point cloud data was collected across multiple construction sites under varying conditions. The dataset encompasses five key excavator activities: Approach, Digging, Dumping, Idle, and Leveling. To assess the effectiveness of the proposed method, the performance of the two-stream CNN-LSTM architecture is compared with that of single-stream CNN-LSTM models on the same RGB and point cloud datasets, separately. The results demonstrate that the proposed multi-stream approach achieved an accuracy of 94.67%, outperforming existing state-of-the-art single-stream models, which achieved 90.67% accuracy for the RGB-based model and 92.00% for the point cloud-based model. These findings underscore the potential of the proposed activity recognition method, making it highly effective for automatic real-time monitoring of excavator activities, thereby laying the groundwork for future integration into digital twin systems for proactive maintenance and intelligent equipment management. Full article

Magnetotelluric (MT) sounding is a crucial technique in mineral exploration. However, MT data are highly susceptible to various types of noise. Traditional data processing methods, which rely on the assumption of signal stationarity, often result in severe distortion when suppressing non-stationary noise. In this study, we propose a novel, adaptive, and less parameter-dependent signal decomposition method for MT signal denoising, based on time–frequency domain analysis and the application of modal decomposition. The method uses Variational Mode Decomposition (VMD) to adaptively decompose the MT signal into several intrinsic mode functions (IMFs), obtaining the instantaneous time–frequency energy distribution of the signal. Subsequently, robust statistical methods are introduced to extract the independent components of each IMF, thereby identifying signal and noise components within the decomposition results. Synthetic data experiments show that our method accurately separates high-amplitude non-stationary interference. Furthermore, it maintains stable decomposition results under various parameter settings, exhibiting strong robustness and low parameter dependency. When applied to field MT data, the method effectively filters out non-stationary noise, leading to significant improvements in both apparent resistivity and phase curves, indicating its practical value in mineral exploration. Full article

Accurate estimation of Interrupted Sampling Repeater Jamming (ISRJ) sampling duration is essential for effective radar anti-jamming. However, in complex electromagnetic environments, the simultaneous presence of suppressive and deceptive jamming, coupled with significant signal overlap in the time–frequency domain, renders ISRJ separation and parameter estimation considerably challenging. To address this challenge, this paper proposes a method utilizing the Fractional Fourier Transform (FRFT) and a Hybrid Modality Fusion Network (HMFN) for ISRJ signal separation and sampling-duration estimation. The proposed method first employs FRFT and a time–frequency mask to separate the ISRJ and target echo from the mixed signal. This process effectively suppresses interference and extracts the ISRJ signal. Subsequently, an HMFN is employed for high-precision estimation of the ISRJ sampling duration, offering crucial parameter support for active electromagnetic countermeasures. Simulation results validate the performance of the proposed method. Specifically, even under strong interference conditions with a Signal-to-Jamming Ratio (SJR) of −5 dB for deceptive jamming and as low as −10 dB for suppressive jamming, the regression model’s coefficient of determination still reaches 0.91. This result clearly demonstrates the method’s robustness and effectiveness in complex electromagnetic environments. Full article

Prognostic markers such as overall survival (OS) and tertiary lymphoid structure (TLS) ratios, alongside diagnostic signatures like primary cancer-type classification, provide critical information for treatment selection, risk stratification, and longitudinal care planning across the oncology continuum. However, extracting these signals solely from sparse, high-dimensional multi-omics data remains a major challenge due to heterogeneity and frequent missingness in patient profiles. To address this challenge, we present SeNMo, a self-normalizing deep neural network trained on five heterogeneous omics layers—gene expression, DNA methylation, miRNA abundance, somatic mutations, and protein expression—along with the clinical variables, that learns a unified representation robust to missing modalities. Trained on more than 10,000 patient profiles across 32 tumor types from The Cancer Genome Atlas (TCGA), SeNMo provides a baseline that can be readily fine-tuned for diverse downstream tasks. On a held-out TCGA test set, the model achieved a concordance index of 0.758 for OS prediction, while external evaluation yielded 0.73 on the CPTAC lung squamous cell carcinoma cohort and 0.66 on an independent 108-patient Moffitt Cancer Center cohort. Furthermore, on Moffitt’s cohort, baseline SeNMo fine-tuned for TLS ratio prediction aligned with expert annotations (p < 0.05) and sharply separated high- versus low-TLS groups, reflecting distinct survival outcomes. Without altering the backbone, a single linear head classified primary cancer type with 99.8% accuracy across the 33 classes. By unifying diagnostic and prognostic predictions in a modality-robust architecture, SeNMo demonstrated strong performance across multiple clinically relevant tasks, including survival estimation, cancer classification, and TLS ratio prediction, highlighting its translational potential for multi-omics oncology applications. Full article

We present mmHSE, a two-stage framework for human skeleton estimation using dual millimeter-Wave (mmWave) Frequency-Modulated Continuous-Wave (FMCW) radar signals. To enable data-driven model design and evaluation, we collect and process over 30,000 range–angle maps from 12 users across three representative indoor environments using a dual-node radar acquisition platform. Leveraging the collected data, we develop a two-stage neural architecture for human skeleton estimation. The first stage employs a dual-branch network with depthwise separable convolutions and self-attention to extract multi-scale spatiotemporal features from dual-view radar inputs. A cross-modal attention fusion module is then used to generate initial estimates of 21 skeletal keypoints. The second stage refines these estimates using a skeletal topology module based on graph convolutional networks, which captures spatial dependencies among joints to enhance localization accuracy. Experiments show that mmHSE achieves a Mean Absolute Error (MAE) of 2.78 cm. In cross-domain evaluations, the MAE remains at 3.14 cm, demonstrating the method’s generalization ability and robustness for non-intrusive human pose estimation from mmWave FMCW radar signals. Full article

Image captioning plays a crucial role in aligning visual content with natural language, serving as a key step toward effective cross-modal understanding. Transformer has become the dominant language model in image captioning. Existing Transformer-based models seldom highlight important features from multiple views in the use of self-attention. In this paper, we propose MFEAM, an innovative network that leverages the multi-view feature enhanced attention. To accurately represent the entangled features of vision and text, the teacher model employs the multi-view feature enhanced attention to guide the student model training through knowledge distillation and model averaging from both visual and textual views. To mitigate the impact of excessive feature enhancement, the student model divides the decoding layer into two groups, which separately process instance features and the relationships between instances. Experimental results demonstrate that MFEAM attains competitive performance on the MSCOCO (Microsoft Common Objects in Context) when trained without leveraging external data. Full article

The rise of virtual production has created an urgent need for both efficient and high-fidelity 3D face generation schemes for cinema and immersive media, but existing methods are often limited by lighting–geometry coupling, multi-view dependency, and insufficient artistic quality. To address this, this study proposes a cross-modal 3D face generation framework based on single-view semantic masks. It utilizes Swin Transformer for multi-level feature extraction and combines with NeRF for illumination decoupled rendering. We utilize physical rendering equations to explicitly separate surface reflectance from ambient lighting to achieve robust adaptation to complex lighting variations. In addition, to address geometric errors across illumination scenes, we construct geometric a priori constraint networks by mapping 2D facial features to 3D parameter space as regular terms with the help of semantic masks. On the CelebAMask-HQ dataset, this method achieves a leading score of SSIM = 0.892 (37.6% improvement from baseline) with FID = 40.6. The generated faces excel in symmetry and detail fidelity with realism and aesthetic scores of 8/10 and 7/10, respectively, in a perceptual evaluation with 1000 viewers. By combining physical-level illumination decoupling with semantic geometry a priori, this paper establishes a quantifiable feedback mechanism between objective metrics and human aesthetic evaluation, providing a new paradigm for aesthetic quality assessment of AI-generated content. Full article

Control valves in nuclear systems operate under high-pressure differentials generating intense transient fluid forces that induce destructive structural vibrations, risking resonance and the valve stem fracture. In this study, computational fluid dynamics (CFD) was employed to characterize the internal flow dynamics of the valve, supported by experiment validation of the fluid model. To account for nonlinear structural effects such as contact and damping, a coupled fluid–structure interaction approach incorporating nonlinear perturbation analysis was applied to evaluate the dynamic response of the valve core assembly under fluid excitation. The results indicate that flow separation, re-circulation, and vortex shedding within the throttling region are primary contributors to structural vibrations. A comparative analysis of stability coefficients, modal damping ratios, and logarithmic decrements under different valve openings revealed that the valve core assembly remains relatively stable overall. However, critical stability risks were identified in the lower-order modal frequency range at 50% and 70% openings. Notably, at a 70% opening, the first-order modal frequency of the valve core assembly closely aligns with the frequency of fluid excitation, indicating a potential for critical resonance. This research provides important insights for evaluating and enhancing the vibration stability and operational safety of control valves under complex flow conditions. Full article

This paper presents a formulation for the dynamic analysis of dam–reservoir–foundation systems, employing a coupled finite element model that integrates displacements and reservoir pressures. An innovative coupled approach, without separating the solid and fluid equations, is proposed to directly solve the single non-symmetrical governing equation for the whole system with non-proportional damping. For the modal analysis, a state–space method is adopted to solve the coupled eigenproblem, and complex eigenvalues and eigenvectors are computed, corresponding to non-stationary vibration modes. For the seismic analysis, a time-stepping method is applied to the coupled dynamic equation, and the stress–transfer method is introduced to simulate the nonlinear behavior, innovatively combining a constitutive joint model and a concrete damage model with softening and two independent scalar damage variables (tension and compression). This formulation is implemented in the computer program DamDySSA5.0, developed by the authors. To validate the formulation, this paper provides the experimental and numerical results in the case of the Cahora Bassa dam, instrumented in 2010 with a continuous vibration monitoring system designed by the authors. The good comparison achieved between the monitoring data and the dam–reservoir–foundation model shows that the formulation is suitable for simulating the modal response (natural frequencies and mode shapes) for different reservoir water levels and the seismic response under low-intensity earthquakes, using accelerograms measured at the dam base as input. Additionally, the dam’s nonlinear seismic response is simulated under an artificial accelerogram of increasing intensity, showing the structural effects due to vertical joint movements (release of arch tensions near the crest) and the concrete damage evolution. Full article

Poultry feces, a critical biomarker for health assessment, requires timely and accurate pathological identification for food safety. Conventional visual-only methods face limitations due to environmental sensitivity and high visual similarity among feces from different diseases. To address this, we propose MMCD (Multimodal Chicken-feces Diagnosis), a ResNet50-based multimodal fusion model leveraging semantic complementarity between images and descriptive text to enhance diagnostic precision. Key innovations include the following: (1) Integrating MASA(Manhattan self-attention)and DSconv (Depthwise Separable convolution) into the backbone network to mitigate feature confusion. (2) Utilizing a pre-trained BERT to extract textual semantic features, reducing annotation dependency and cost. (3) Designing a lightweight Gated Cross-Attention (GCA) module for dynamic multimodal fusion, achieving a 41% parameter reduction versus cross-modal transformers. Experiments demonstrate that MMCD significantly outperforms single-modal baselines in Accuracy (+8.69%), Recall (+8.72%), Precision (+8.67%), and F1 score (+8.72%). It surpasses simple feature concatenation by 2.51–2.82% and reduces parameters by 7.5M and computations by 1.62 GFLOPs versus the base ResNet50. This work validates multimodal fusion’s efficacy in pathological fecal detection, providing a theoretical and technical foundation for agricultural health monitoring systems. Full article

Rolling element bearings serve as critical transmission components in industrial automation systems, yet their fault signatures are susceptible to interference from strong background noise, complex operating conditions, and nonlinear impact characteristics. Addressing the limitations of conventional methods in adaptive parameter optimization and weak feature enhancement, this paper proposes an innovative diagnostic framework integrating Improved Goose optimized Variational Mode Decomposition (IGOOSE-VMD), RobustICA, and CYCBD. First, to mitigate modal aliasing issues caused by empirical parameter dependency in VMD, we fuse a refraction-guided reverse learning mechanism with a dynamic mutation strategy to develop the IGOOSE. By employing an energy-feature-driven fitness function, this approach achieves synergistic optimization of the mode number and penalty factor. Subsequently, a multi-channel observation model is constructed based on optimal component selection. Noise interference is suppressed through the robust separation capabilities of RobustICA, while CYCBD introduces cyclostationarity-based prior constraints to formulate a blind deconvolution operator with periodic impact enhancement properties. This significantly improves the temporal sparsity of fault-induced impact components. Experimental results demonstrate that, compared to traditional time–frequency analysis techniques (e.g., EMD, EEMD, LMD, ITD) and deconvolution methods (including MCKD, MED, OMEDA), the proposed approach exhibits superior noise immunity and higher fault feature extraction accuracy under high background noise conditions. Full article

Pulsed radiofrequency (PRF) current applied to peripheral nerves is a modality used in interventional pain medicine, but its underlying mechanisms remain unclear. This study aimed to investigate whether ex vivo exposure of human monocytic THP-1 cells to PRF current or to heat induces β-endorphin production. Methods: THP-1 cells were exposed to PRF current for 15 min or incubated at elevated temperatures (42 °C to 50 °C) for 3 or 15 min. Flow cytometry was used to assess cell viability, and β-endorphin concentrations in culture supernatants were quantified by ELISA. In a separate experiment, cells were stimulated with lipopolysaccharide (LPS) to compare its effects on β-endorphin release. Results: A 3 min exposure to temperatures ≥ 46 °C reduced THP-1 cell viability, whereas a 15 min exposure to PRF current or to heat at 42 °C did not impair viability. Both PRF current and mild heat significantly enhanced β-endorphin release. β-Endorphin levels in the supernatant of LPS-stimulated cells were comparable to those of cells exposed to PRF current. Conclusions: Ex vivo application of PRF current or mild heat enhanced β-endorphin production from THP-1 cells without significant cytotoxicity. These preliminary findings warrant further investigation using primary human monocytes and in vivo models to assess therapeutic potential. Full article

Skin cancer is one of the most prevalent forms of cancer worldwide, and early and accurate diagnosis critically impacts patient outcomes. Given the sensitive nature of medical data and its fragmented distribution across institutions (data silos), privacy-preserving collaborative learning is essential to enable knowledge-sharing without compromising patient confidentiality. While federated learning (FL) offers a promising solution, existing methods struggle with heterogeneous and missing modalities across institutions, which reduce the diagnostic accuracy. To address these challenges, we propose an effective and flexible Personalized Multimodal Federated Learning framework (PMM-FL), which enables efficient cross-client knowledge transfer while maintaining personalized performance under heterogeneous and incomplete modality conditions. Our study contains three key contributions: (1) A hierarchical aggregation strategy that decouples multi-module aggregation from local deployment via global modular-separated aggregation and local client fine-tuning. Unlike conventional FL (which synchronizes all parameters in each round), our method adopts a frequency-adaptive synchronization mechanism, updating parameters based on their stability and functional roles. (2) A multimodal fusion approach based on multitask learning, integrating learnable modality imputation and attention-based feature fusion to handle missing modalities. (3) A custom dataset combining multi-year International Skin Imaging Collaboration(ISIC) challenge data (2018–2024) to ensure comprehensive coverage of diverse skin cancer types. We evaluate PMM-FL through diverse experiment settings, demonstrating its effectiveness in heterogeneous and incomplete modality federated learning settings, achieving 92.32% diagnostic accuracy with only a 2% drop in accuracy under 30% modality missingness, with a 32.9% communication overhead decline compared with baseline FL methods. Full article

Background: Cochlear implantation is widely used to provide auditory rehabilitation to individuals with severe-to-profound sensorineural hearing loss. However, electrode insertion during cochlear implantation leads to inner ear trauma, damage to sensory structures, and consequently, loss of residual hearing. There is very limited information regarding the target proteins involved in electrode insertion trauma (EIT) following cochlear implantation. Methods: The aim of our study was to identify target proteins and host molecular pathways involved in cochlear damage following EIT utilizing the iTRAQ™ (isobaric tags for relative and absolute quantification) technique using our ex vivo model. The organ of Corti (OC) explants were dissected from postnatal day 3 rats and subjected to EIT or left untreated (control). The proteins were extracted, labelled, and subjected to ultra-high performance liquid chromatography–tandem mass spectrometry. Results: We identified distinct molecular pathways involved in EIT-induced cochlear damage. Confocal microscopy confirmed the expression of these identified proteins in OC explants subjected to EIT. By separating the apical, middle, and basal cochlear turns, we deciphered a topographic array of host molecular pathways that extend from the base to the apex of the cochlea, which are activated post-trauma following cochlear implantation. Conclusions: The identification of target proteins involved in cochlear damage will provide novel therapeutic targets for the development of effective treatment modalities for the preservation of residual hearing in implanted individuals. Full article

Small and medium-sized enterprises (SMEs) face growing challenges in optimizing their sustainable supply chains because of fragmented logistics data and changing regulatory requirements. In particular, globally operating manufacturing SMEs often lack suitable tools, resulting in manual data collection and making reliable accounting and benchmarking of transport emissions in lifecycle assessments (LCAs) time-consuming and difficult to scale. This paper introduces a novel hybrid AI-supported knowledge graph (KG) which combines large language models (LLMs) with graph-based optimization to automate industrial supply chain route enrichment, completion, and emissions analysis. The proposed solution automatically resolves transportation gaps through generative AI and programming interfaces to create optimal routes for cost, time, and emission determination. The application merges separate routes into a single multi-modal network which allows users to evaluate sustainability against operational performance. A case study shows the capabilities in simplifying data collection for emissions reporting, therefore reducing manual effort and empowering SMEs to align logistics decisions with Industry 5.0 sustainability goals. Full article