Search Results (1,286)

In practical applications, intelligent diagnostic methods for actuator-integrated gearboxes in industrial driving systems encounter challenges such as the scarcity of fault samples and variable operating conditions, which undermine diagnostic accuracy. This paper introduces a multi-layer feature fusion meta-learning (MLFFML) approach to address fault diagnosis problems in cross-condition scenarios with class imbalance. First, meta-training is performed to develop a mature fault diagnosis model on the source domain, obtaining cross-domain meta-knowledge; subsequently, meta-testing is conducted on the target domain, extracting meta-features from limited fault samples and abundant healthy samples to rapidly adjust model parameters. For data augmentation, this paper proposes a frequency-domain weighted mixing (FWM) method that preserves the physical plausibility of signals while enhancing sample diversity. Regarding the feature extractor, this paper integrates shallow and deep features by replacing the first layer of the feature extraction module with a dual-stream wavelet convolution block (DWCB), which transforms actuator vibration or acoustic signals into the time-frequency space to flexibly capture fault characteristics and fuses information from both amplitude and phase aspects; following the convolutional network, an encoder layer of the Transformer network is incorporated, containing multi-head self-attention mechanisms and feedforward neural networks to comprehensively consider dependencies among different channel features, thereby achieving a larger receptive field compared to other methods for actuation system monitoring. Furthermore, this paper experimentally investigates cross-modal scenarios where vibration signals exist in the source domain while only acoustic signals are available in the target domain, specifically validating the approach on industrial actuator assemblies. Full article

The capability to discriminate among nuclear fuel properties is essential for a successful nuclear safeguard and security program. Accurate nuclear material identification is hindered due to challenges such as differing levels of enrichments, weak radiation signals in the case of fresh nuclear fuel, and complex self-shielding effects. This study explores the application of supervised machine learning algorithms to digitized radiation detector data for classifying signatures of special nuclear materials. Three scintillation detectors, an EJ-309 liquid scintillator, a CLYC crystal scintillator, and an EJ-276 plastic scintillator, were used to measure gamma-ray and neutron data from special nuclear material at the National Criticality Experiments Research Center (NCERC) at the National Nuclear Security Site (NNSS), at Nevada, USA. Radiation detector pulse data was extracted from the collected digitized data and applied to three separate supervised learning models: Random Forest, XGBoost, and a feedforward Deep Neural Network, chosen for their wide-spread use and distinct data ingest and processing analytics. Through model refinement, such as adding an additional parameter feature, an accuracy of greater than 95% was achieved. Analysis on model parameter feature importance revealed Countrate, which is the overall gamma-ray and neutron incidents for each detector, was the most influential parameter and essential to include for improved classification. Initial model versions not including the Countrate parameter feature failed to classify. Supervised learning models allow for measured gamma-ray and neutron pulse data to be used to develop effective identification and discrimination between material compositions of different fuel assemblies. The study demonstrated that traditional pulse shape parameters alone were insufficient for discriminating between special nuclear materials; the addition of Countrate markedly improved model accuracy but all models were heavily dependent on this specific feature, thus illustrating the need for alternative, more distinct parameter features. The machine learning development framework captured in this study will be beneficial for future applications in discriminating between different fuel enrichments and additives such as burnable poisons. Full article

Basal-like breast cancer (BLBC) is associated with poor prognosis, high recurrence rates, and limited therapeutic options, largely due to its molecular heterogeneity and complexity, which include epigenetic alterations. This study investigated epigenetic regulatory networks in BLBC by analyzing DNA methylation in distal cis-regulatory regions and its impact on genes, transcription factors (TFs), and microRNAs (miRNAs) expression. Data from TCGA were processed using the ELMER and DESeq2 tools to identify differentially methylated regions and differentially expressed genes, TFs, and miRNAs. The FANMOD algorithm was used to identify the regulatory interactions uncovering the feed-forward loops (FFLs). The analysis identified 110 TF-mediated FFLs, 43 miRNA-mediated FFLs, and five composite FFLs, involving 18 hypermethylated and 32 hypomethylated genes, eight upregulated and nine downregulated TFs, and 21 upregulated and seven downregulated miRNAs. The TF-mediated FFLs major regulators involved the AR, EBF1, FOS, FOXM1, and TEAD4 TFs, while key miRNAs were miR-3662, miR-429, and miR-4434. Enriched pathways involved cAMP, ErbB, FoxO, p53, TGF-beta, Rap1, and Ras signaling. Differences in hallmark gene set categories reflected distinct methylation and miRNA expression profiles. Overall, this integrative analysis mapped the intricate epigenetic landscape of BLBC, emphasizing the role of FFLs as regulatory motifs that integrate DNA methylation, TFs, and miRNAs in orchestrating disease’s development and progression and offering potential targets for future diagnostic and therapeutic strategies. Full article

In the present study, the changes in the groundwater quality in a Hungarian settlement, Báránd, were examined, nine years after the construction of a sewerage network. The sewerage network in the study area was completed in 2014, with a household connection rate exceeding 97% in 2023. In the summer of 2023, water samples were taken from 37 dug groundwater wells. Changes in the water quality were assessed using three water quality indicators (the Water Quality Index (WQI), Contamination degree (Cd), and Canadian Council of Ministers of the Environment Water Quality Index (CCME WQI)) and geographic information (GIS), data visualization systems, and artificial intelligence (AI). During the evaluation of the quality of the groundwater, eight water chemical parameters were used (pH, EC, NH₄⁺, NO₂⁻, NO₃⁻, PO₄³⁻, COD, Na⁺). Based on interpolated maps and water quality indices, it was established that while an increasing portion of the area exhibits adequate or good water quality compared to the pre-sewerage period, a deterioration has occurred relative to recent years. Even nine years after the sewerage network construction, elevated concentrations of inorganic nitrogen forms and organic matter persist, indicating the continued presence of accumulated pollutants, as confirmed by all three water quality indicators to varying degrees and spatial patterns. The interactive data visualization and cloud-based sharing of the data of the water quality geodatabase were made freely available with the help of Tableau Public. A Feed-Forward Neural Network (FFNN) was developed to predict the groundwater quality, estimating the water quality statuses of three water quality indicators based on water chemistry parameters. The results showed that the applied training algorithms and activation functions proved to be the most effective in the case of different network structures. The most accurate prediction of the WQI and CCME WQI indicators was provided by the Bayesian control algorithm (trainbr), which achieved the lowest mean-squared error (RMSE_WQI = 0.1205, RMSE_{CCME WQI} = 0.1305) and the highest determination coefficient (R²_WQI = 0.9916, R²_{CCME WQI} = 0.9838). For the Cd index, the accuracy of the model was lower (RMSE = 0.1621, R² = 0.9714), suggesting that this indicator is more difficult to predict. With regard to our study, it should be emphasized that data visualization is a particularly practical tool for the post-processing of spatial monitoring data, as it is suitable for displaying information in an intuitive, visual form, for discovering spatial patterns and relationships, and for performing real-time analyses. AI is expected to further increase visualization efficiency in the future, enabling the rapid processing of large amounts of data and spatial databases, as well as the identification of complex patterns. Full article

Based on the analysis of solar array current data from a certain MEO-orbiting satellite, this paper reveals its short-period fluctuation characteristics and underlying mechanisms. The study finds that when solar panels face the sun during the light period, the output current exhibits significant short-period fluctuations in addition to being influenced by long-period factors such as sun–earth distance, incident light intensity changes, and space irradiation attenuation. Through theoretical analysis, we first confirm that the root cause of these short-period variations is the temperature change in the shunt circuit caused by load fluctuations, which in turn affects the output current characteristics. Unlike traditional methods that use static characteristic factors such as incident angles, this paper innovatively proposes using load current as a key characteristic factor. For asymmetric solar panel fault scenarios, load current, time phase, and fault-wing output current are used as characteristic factors to adaptively predict the current of normal wings. Meanwhile, feedforward neural network (FNN), Recurrent Neural Network (RNN), and long short-term memory (LSTM) are used for output current prediction. The experimental results show that these methods can accurately capture the short-period fluctuations caused by load mutations and adapt to the fluctuation trend of the normal wing during the prediction of current changes in the faulty wing. It is worth noting that, limited by the short-period fluctuation prediction scenario, the inherent advantage of LSTM in long-sequence prediction is not fully reflected. Full article

Efficient bandwidth allocation in 5G networks is essential for optimizing network performance and ensuring high quality of service (QoS), particularly in unmanned aerial vehicle (UAV) communication systems. The dynamic nature of UAV networks presents challenges in managing fluctuating QoS levels, necessitating intelligent bandwidth allocation strategies. This study investigates the effectiveness of two machine learning (ML) models, least square gradient boosting (LSGB) and a Bayesian regularization feedforward neural network (BRFFNN), in predicting bandwidth allocation for UAV ground station (UAV-GS) communication under 5G specifications. Using a simulation-based approach, the study evaluates UAV bandwidth allocation under two movement patterns: circular and random. The QoS metrics considered include the packet delivery ratio (PDR), delay, and throughput. The results demonstrate that the BRFFNN outperforms LSGB, particularly in circular UAV movement, achieving a 100% PDR, a 0.00773 ms delay, and a 3.232 million packets per second (pps) throughput. These findings suggest that ML models, particularly the BRFFNN, can significantly enhance bandwidth allocation strategies in 5G UAV-GS communication systems, improving overall network efficiency and QoS. This study provides valuable insights into ML-driven bandwidth allocation, emphasizing the BRFFNN as a superior approach for enhancing QoS in 5G UAV-GS networks. In the context of 5G UAV-GS bandwidth allocation, this study applies the BRFFNN in a novel way and demonstrates its superiority over tree-based models such as LSGB. In contrast to earlier research that concentrated on static or traditional allocation techniques, our method achieves State-of-the-Art QoS by dynamically predicting bandwidth under actual UAV movement scenarios. Full article

Existing nighttime visibility enhancement methods primarily focus on improving overall brightness under low-light conditions. However, nighttime road images are also affected by glare, glow, and flare from complex light sources such as streetlights and headlights, making it challenging to suppress locally overexposed regions and recover fine details. To address these challenges, we propose a Nighttime Road Glare Suppression Network (NRGS-Net) for glare removal and detail restoration. Specifically, to handle diverse glare disturbances caused by the uncertainty in light source positions and shapes, we designed a gated positional attention (GPA) module that integrates positional encoding with local contextual information to guide the network in accurately locating and suppressing glare regions, thereby enhancing the visibility of affected areas. Furthermore, we introduced an improved Uformer backbone named LCAtransformer, in which the downsampling layers adopt efficient depthwise separable convolutions to reduce computational cost while preserving critical spatial information. The upsampling layers incorporate a residual PixelShuffle module to achieve effective restoration in glare-affected regions. Additionally, channel attention is introduced within the Local Context-Aware Feed-Forward Network (LCA-FFN) to enable adaptive adjustment of feature weights, effectively suppressing irrelevant and interfering features. To advance the research in nighttime glare suppression, we constructed and publicly released the Night Road Glare Dataset (NRGD) captured in real nighttime road scenarios, enriching the evaluation system for this task. Experiments conducted on the Flare7K++ and NRGD, using five evaluation metrics and comparing six state-of-the-art methods, demonstrate that our method achieves superior performance in both subjective and objective metrics compared to existing advanced methods. Full article

Artificial intelligence (AI) has emerged as a transformative technology in the evolution of massive multiple-input multiple-output (ma-MIMO) systems, positioning them as a cornerstone for sixth-generation (6G) wireless networks. Despite their significant potential, ma-MIMO systems face critical challenges at the receiver end, particularly in signal detection under high-dimensional and noisy environments. To address these limitations, this paper proposes MIMONet, a novel deep learning (DL)-based MIMO detection framework built upon a lightweight and optimized feedforward neural network (FFNN) architecture. MIMONet is specifically designed to achieve a balance between detection performance and complexity by optimizing the neural network architecture for MIMO signal detection tasks. Through extensive simulations across multiple MIMO configurations, the proposed MIMONet detector consistently demonstrates superior bit error rate (BER) performance. It achieves a notably lower error rate compared to conventional benchmark detectors, particularly under moderate to high signal-to-noise ratio (SNR) conditions. In addition to its enhanced detection accuracy, MIMONet maintains significantly reduced computational complexity, highlighting its practical feasibility for advanced wireless communication systems. These results validate the effectiveness of the MIMONet detector in optimizing detection accuracy without imposing excessive processing burdens. Moreover, the architectural flexibility and efficiency of MIMONet lay a solid foundation for future extensions toward large-scale ma-MIMO configurations, paving the way for practical implementations in beyond-5G (B5G) and 6G communication infrastructures. Full article

The organization of consciousness is described through increasingly rich theoretical models. We review evidence that working memory capacity—essential to generating consciousness in the cerebral cortex—is supported by dual limbic memory systems. These dorsal (Papez) and ventral (Yakovlev) limbic networks provide the basis for mnemonic processing and prediction in the dorsal and ventral divisions of the human neocortex. Empirical evidence suggests that the dorsal limbic division is (i) regulated preferentially by excitatory feedforward control, (ii) consolidated by REM sleep, and (iii) controlled in waking by phasic arousal through lemnothalamic projections from the pontine brainstem reticular activating system. The ventral limbic division and striatum, (i) organizes the inhibitory neurophysiology of NREM to (ii) consolidate explicit memory in sleep, (iii) operating in waking cognition under the same inhibitory feedback control supported by collothalamic tonic activation from the midbrain. We propose that (i) these dual (excitatory and inhibitory) systems alternate in the stages of sleep, and (ii) in waking they must be balanced—at criticality—to optimize the active inference that generates conscious experiences. Optimal Bayesian belief updating rests on balanced feedforward (excitatory predictive) and feedback (inhibitory corrective) control biases that play the role of prior and likelihood (i.e., sensory) precision. Because the excitatory (E) phasic arousal and inhibitory (I) tonic activation systems that regulate these dual limbic divisions have distinct affective properties, varying levels of elation for phasic arousal (E) and anxiety for tonic activation (I), the dual control systems regulate sleep and consciousness in ways that are adaptively balanced—around the entropic nadir of E–I criticality—for optimal self-regulation of consciousness and psychological health. Because they are emotive as well as motive control systems, these dual systems have unique qualities of feeling that may be registered as subjective experience. Full article

Membrane transporters play a crucial role in any cell. Identifying the substrates they translocate across membranes is important for many fields of research, such as metabolomics, pharmacology, and biotechnology. In this study, we leverage recent advances in deep learning, such as amino acid sequence embeddings with protein language models (pLMs), highly accurate 3D structure predictions with AlphaFold 2, and structure-encoding 3Di sequences from FoldSeek, for predicting substrates of membrane transporters. We test new deep learning features derived from both sequence and structure, and compare them to the previously best-performing protein encodings, which were made up of amino acid k-mer frequencies and evolutionary information from PSSMs. Furthermore, we compare the performance of these features either using a previously developed SVM model, or with a regularized feedforward neural network (FNN). When evaluating these models on sugar and amino acid carriers in A. thaliana, as well as on three types of ion channels in human, we found that both the DL-based features and the FNN model led to a better and more consistent classification performance compared to previous methods. Direct encodings of 3D structures with Foldseek, as well as structural embeddings with ProstT5, matched the performance of state-of-the-art amino acid sequence embeddings calculated with the ProtT5-XL model when used as input for the FNN classifier. Full article

Power distribution systems are witnessing a growing deployment of distributed, inverter-based renewable resources such as solar generation. This poses certain challenges such as rapid voltage fluctuations due to the intermittent nature of renewables. Volt-Var control (VVC) methods have been proposed to utilize the ability of inverters to supply or consume reactive power to mitigate fast voltage fluctuations. These methods usually require a detailed power network model including topology and impedance data. However, network models may be difficult to obtain. Thus, it is desirable to develop a model-free method that obviates the need for the network model. This paper proposes a novel model-free cooperative control method to perform voltage regulation and reduce inverter aging in power distribution systems. This method assumes the existence of time-series voltage and load data, from which the relationship between voltage and nodal power injection is derived using a feedforward artificial neural network (ANN). The node voltage sensitivity versus reactive power injection can then be calculated, based on which a cooperative control approach is proposed for mitigating voltage fluctuation. The results obtained for a modified IEEE 13-bus system using the proposed method have shown its effectiveness in mitigating fast voltage variation due to PV intermittency. Moreover, a comparative analysis between model-free and model-based methods is provided to demonstrate the feasibility of the proposed method. Full article

Background: Unlike most cancers, breast cancer poses a persistent risk of distant recurrence—often years after initial treatment—making long-term risk stratification uniquely challenging. Current tools fall short in predicting late metastatic events, particularly for early-stage patients. Methods: We present an interpretable machine learning (ML) pipeline to predict distant recurrence-free survival at 5, 10, and 15 years, integrating Bayesian network-based causal feature selection, deep feed-forward neural network models (DNMs), and SHAP-based interpretation. Using electronic health record (EHR)-based clinical data from over 6000 patients, we first applied the Markov blanket and interactive risk factor learner (MBIL) to identify minimally sufficient predictor subsets. These were then used to train optimized DNM classifiers, with hyperparameters tuned via grid search and benchmarked against models from 10 traditional ML methods and models trained using all predictors. Results: Our best models achieved area under the curve (AUC) scores of 0.79, 0.83, and 0.89 for 5-, 10-, and 15-year predictions, respectively—substantially outperforming baselines. MBIL reduced input dimensionality by over 80% without sacrificing accuracy. Importantly, MBIL-selected features (e.g., nodal status, hormone receptor expression, tumor size) overlapped strongly with top SHAP contributors, reinforcing interpretability. Calibration plots further demonstrated close agreement between predicted probabilities and observed recurrence rates. The percentage performance improvement due to grid search ranged from 25.3% to 60%. Conclusions: This study demonstrates that combining causal selection, deep learning, and grid search improves prediction accuracy, transparency, and calibration for long-horizon breast cancer recurrence risk. The proposed framework is well-positioned for clinical use, especially to guide long-term follow-up and therapy decisions in early-stage patients. Full article

This study proposes a deep learning-based method to improve overlay alignment precision in laser direct writing systems. Alignment errors arise from multiple sources in nanoscale processes, including optical aberrations, mechanical drift, and fiducial mark imperfections. A significant portion of the residual alignment error stems from the interpretation of mark coordinates by the vision system and algorithms. Here, we developed a convolutional neural network (CNN) model to predict the coordinates calculation error of 66,000 sets of computer-generated defective crosshair marks (simulating real fiducial mark imperfections). We compared 14 neural network architectures (8 CNN variants and 6 feedforward neural network (FNN) configurations) and found a well-performing, simple CNN structure achieving a mean squared error (MSE) of 0.0011 on the training sets and 0.0016 on the validation sets, demonstrating 90% error reduction compared to the FNN structure. Experimental results on test datasets showed the CNN’s capability to maintain prediction errors below 100 nm in both X/Y coordinates, significantly outperforming traditional FNN approaches. The proposed method’s success stems from the CNN’s inherent advantages in local feature extraction and translation invariance, combined with a simplified network architecture that prevents overfitting while maintaining computational efficiency. This breakthrough establishes a new paradigm for precision enhancement in micro–nano optical device fabrication. Full article

In the task of PolSAR image classification, effectively utilizing convolutional neural networks and vision transformer models with limited labeled data poses a critical challenge. This article proposes a novel method for PolSAR image classification that combines channel-wise convolutional local perception, detachable self-attention, and a residual feedforward network. Specifically, the proposed method comprises several key modules. In the channel-wise convolutional local perception module, channel-wise convolution operations enable accurate extraction of local features from different channels of PolSAR images. The local residual connections further enhance these extracted features, providing more discriminative information for subsequent processing. Additionally, the detachable self-attention mechanism plays a pivotal role: it facilitates effective interaction between local and global information, enabling the model to comprehensively perceive features across different scales, thereby improving classification accuracy and robustness. Subsequently, replacing the conventional feedforward network with a residual feedforward network that incorporates residual structures aids the model in better representing local features, further enhances the capability of cross-layer gradient propagation, and effectively alleviates the problem of vanishing gradients during the training of deep networks. In the final classification stage, two fully connected layers with dropout prevent overfitting, while softmax generates predictions. The proposed method was validated on the AIRSAR Flevoland, RADARSAT-2 San Francisco, and RADARSAT-2 Xi’an datasets. The experimental results demonstrate that the proposed method can attain a high level of classification performance even with a limited amount of labeled data, and the model is relatively stable. Furthermore, the proposed method has lower computational costs than comparative methods. Full article

The remote sensing classification of urban functional zones provides scientific support for urban planning, land resource optimization, and ecological environment protection. However, urban functional zone classification encounters significant challenges in accuracy and efficiency due to complicated image structures, ambiguous critical features, and high computational complexity. To tackle these challenges, this work proposes a novel XT-SECA algorithm employing a strengthened efficient channel attention mechanism (SECA) to integrate the feature-extraction XGBoost branch and the feature-enhancement Transformer feedforward branch. The SECA optimizes the feature-fusion process through dynamic pooling and adaptive convolution kernel strategies, reducing feature confusion between various functional zones. XT-SECA is characterized by sufficient learning of complex image structures, effective representation of significant features, and efficient computational performance. The Futian, Luohu, and Nanshan districts in Shenzhen City are selected to conduct urban functional zone classification by XT-SECA, and they feature administrative management, technological innovation, and commercial finance functions, respectively. XT-SECA can effectively distinguish diverse functional zones such as residential zones and public management and service zones, which are easily confused by current mainstream algorithms. Compared with the commonly adopted algorithms for urban functional zone classification, including Random Forest (RF), Long Short-Term Memory (LSTM) network, and Multi-Layer Perceptron (MLP), XT-SECA demonstrates significant advantages in terms of overall accuracy, precision, recall, F1-score, and Kappa coefficient, with an accuracy enhancement of 3.78%, 42.86%, and 44.17%, respectively. The Kappa coefficient is increased by 4.53%, 51.28%, and 52.73%, respectively. Full article