Search Results (255)

Diabetic retinopathy (DR) is a leading cause of vision impairment and permanent blindness worldwide, requiring accurate and automated systems for multi-grade severity classification. However, standard Convolutional Neural Networks (CNNs) often struggle to capture fine, high-frequency microvascular patterns critical for diagnosis. This study proposes a Robust Intelligent CNN Model (RICNN) that integrates Gabor-based feature extraction with deep learning to improve DR classification. Specifically, Gabor filters are applied during preprocessing to extract orientation- and frequency-sensitive texture features, which are transformed into feature maps and concatenated with CNN feature representations at the fully connected layer (feature-level fusion). The model also incorporates the Synthetic Minority Oversampling Technique (SMOTE) for data balancing and the Adam optimizer for efficient convergence. This integration enhances sensitivity to microvascular structures such as microaneurysms and hemorrhages. The proposed RICNN was evaluated on the Messidor dataset (1200 images) across four severity levels: Mild, Moderate, Severe, and Proliferative DR. The model achieved an accuracy of 89%, a precision of 88.75%, a recall of 89%, and an F1-score of 89%, with AUCs of 97% for Severe DR and 99% for Proliferative DR. Comparative analysis confirms that the proposed texture-aware Gabor enhancement significantly outperforms LBP and Color Histogram approaches, indicating its potential for reliable clinical decision support. Full article

With the rapid development of Intelligent Transportation Systems (ITSs), traffic prediction, a crucial component of ITSs, has garnered growing scholarly attention. The appli-cation of deep learning into traffic prediction has emerged as a prominent research direction, especially amid the rapid advancement of pretrained large language models (LLMs), which offer substantial benefits in time-series analysis through cross-modal knowledge transfer. In response to this advancement, this study introduces an innovative model for traffic flow prediction, designated as WGLLM. To capture spatiotemporal characteristics inherent in traffic flow data, this model incorporates a sequence embedding layer constructed on the stationary wavelet transform (SWT) and long short-term memory (LSTM), in conjunction with a spatial embedding layer founded on graph convolutional networks (GCNs). Additionally, a fully connected layer is utilized to integrate embeddings into the LLMs for comprehensive global dependency analysis. To verify the effectiveness of the proposed approach, experiments were carried out on two real traffic flow datasets. The experimental results demonstrate that WGLLM achieves superior predictive performance compared to multiple mainstream baseline models, accompanied by a significant enhancement in prediction accuracy. Full article

Acute Lymphoblastic Leukemia (ALL) is a highly prevalent hematological malignancy, especially in children, for whom precise and prompt subtype identification is essential to establish suitable treatment protocols. Current deep learning-based computer-aided diagnosis (CAD) methods for identifying ALL are hindered by numerous drawbacks, such as a dependence on solely spatial feature depictions, elevated feature dimensions, computationally extensive deep learning architectures, inadequate multi-layer feature utilization, and poor interpretability. This paper introduces LUMINA-Net, a custom, lightweight, and interpretable deep learning CAD for the automated identification and subtype diagnosis of ALL using microscopic blood smear pictures. LUMINA-Net makes four principal contributions: first, it integrates a self-attention module within a lightweight custom Convolution Neural Network (CNN) to effectively capture long-range spatial relationships across clinically pertinent cytological patterns while preserving a compact design. Second, it employs a Discrete Wavelet Transform (DWT)-based wavelet pooling layer that decreases feature dimensions by up to 96.875% while enhancing the obtained depictions with spatial-spectral information. Third, it utilizes a multi-layer feature fusion strategy that combines wavelet-pooled features from two deep layers with a third fully connected layer to create a discriminating multi-scale feature vector. Fourth, it incorporates Gradient-weighted Class Activation Mapping as a dedicated explainability process to furnish clinicians with apparent visual explanations for each classification decision. Withoit the need for image enhancement or segmentation preprocessing, LUMINA-Net outperforms the competing state-of-the-art methods on the same dataset, achieving a peak accuracy of 99.51%, specificity of 99.84%, and sensitivity of 99.51% on the publicly available Kaggle ALL dataset. This demonstrates that LUMINA-Net has the potential to be a dependable, effective, and clinically interpretable CAD tool for ALL diagnosis. Full article

Background: Genome-scale metabolic models (GEMs) are foundational tools for systems biology, enabling quantitative interrogation of human metabolism across physiological and pathological states. However, many legacy reconstructions exhibit heterogeneous identifier usage, incomplete pathway integration, and limited thermodynamic refinement, constraining reproducibility, interoperability, and translational applicability. Methods: We present EHMN 2026, an update of the Edinburgh Human Metabolic Network. The reconstruction was refined through systematic identifier reconciliation using MetaNetX and ChEBI mappings, duplicate reaction consolidation, thermodynamic directionality assessment, and structured pathway annotation via Reactome. The final model was encoded in Systems Biology Markup Language (SBML) Level 3 Version 2 with the Flux Balance Constraints (FBC2) package, ensuring explicit gene–protein–reaction (GPR) representation and compatibility with modern constraint-based modelling toolchains. Results: EHMN 2026 comprises 11 compartments, 14,321 metabolites (species), and 22,642 reactions, supported by 3996 gene products. Of all reactions, 9638 (42.6%) contain GPR associations, linking metabolic transformations to 2887 unique Ensembl gene identifiers (ENSG). Pathway integration yielded 2194 unique Reactome identifiers, providing structured pathway-level organisation of metabolic functions. Thermodynamic refinement reduced infeasible energy-generating cycles and improved reaction directionality coherence while preserving global network connectivity. The reconstruction is fully SBML-compliant and portable across major modelling platforms. Compared with Recon3D and Human1, EHMN 2026 uniquely combines native Reactome reaction-level annotation, systematic MetaNetX identifier harmonisation, documented thermodynamic cycle elimination (37 cycles, 0 remaining), and an 11-compartment architecture supporting organelle-specific modelling—features designed for QSP and multi-layer integration applications. Conclusions: EHMN 2026 delivers a rigorously harmonised, thermodynamically refined, and pathway-annotated human metabolic reconstruction with enhanced annotation depth and standards-based interoperability. By combining genome-scale coverage with structured gene and pathway integration, the model establishes a robust computational backbone for reproducible metabolic analysis and provides a scalable foundation for future multi-layer systems pharmacology and integrative modelling frameworks. Full article

User behavior anomaly detection plays a vital role in network security for identifying malicious access and abnormal activities in high-dimensional internet user behavior data. Although Transformer architectures have been widely adopted in anomaly detection tasks, and their integration with Variational Autoencoders (VAEs) has often been used to further improve detection accuracy, existing integration methods have failed to effectively balance global feature dependency modeling and generative data distribution learning. This results in limited capability in identifying complex anomalous patterns. To address this issue, this paper proposes DF-TransVAE, a novel deeply integrated framework that advances the integration of a Transformer and a VAE for supervised anomaly detection. The framework first fuses global contextual representations from the Transformer encoder with original input features, then maps the fused representation into the latent space via the VAE encoder. A cross-attention mechanism is introduced as the core of deep integration, enabling dynamic, bidirectional interaction between the fused features and latent variables to enhance information fusion. Lastly, a fully connected classifier equipped with residual connections outputs anomaly probabilities for supervised binary classification. Experimental results on two public datasets demonstrate that the proposed framework achieves better performance than existing deep learning methods in terms of accuracy, precision, recall, and F1-score, particularly in detecting complex anomalous patterns. Our results indicate that the deep integration mechanism we propose effectively addresses the limitations of conventional Transformer–VAE combinations. Full article

This study demonstrates that high-performance work systems (HPWSs)—encompassing selective staffing, extensive training, performance incentives, and employee participation—significantly enhance the dimensions of psychological empowerment, including meaning, competence, self-determination, and impact, thereby fully mediating a stronger stay intention. Leader–member exchange (LMX) amplifies both HPWS-to-empowerment and empowerment-to-stay intention pathways via dyadic trust, while team–member exchange (TMX) strengthens initial resource uptake. Theoretically, based on social exchange theory, the results enhance relational exchange frameworks by emphasizing LMX’s superior function over TMX in high-contact situations, positioning empowerment as the critical mechanism connecting HRM practices to loyalty in service sectors. The cornerstones of sustainable hospitality development include employee-centric strategies that foster empowerment through value-aligned jobs, certifications that enhance skills, independent guest service decision-making, and feedback loops that transform HPWSs into long-lasting retention engines by integrating TMX peer networks for cooperative support with LMX through individualized coaching and feedback that fosters trust. By reducing attrition, stabilizing talent pipelines, and stimulating service quality innovation, these tactics promote robust operations and sustained competitiveness. Full article

Understanding the intimate interactions between plants and their microbiota at the cellular level is essential for unlocking the full potential of plant holobionts in agricultural systems. Traditional bulk and microbial community-level sequencing approaches reveal broad community patterns but fail to resolve how distinct plant cell types interact with or regulate microbial colonization, as well as the diverse antagonistic and synergistic interactions and responses existing between various microbial populations. Recent advances in single-cell and spatial multi-omics have transformed our understanding of plant cell identities as well as gene regulatory programs and their dynamic regulation in response to environmental stresses and plant development. In this review, we highlight the single-cell discoveries that uncover the plant cell-type-specific microbial perception, immune activation, and symbiotic differentiation, particularly in roots, nodules, and leaves. We further discuss how integrating transcriptomic, epigenomic, and spatial data can reconstruct multilayered interaction networks that connect plant cell-type-specific regulatory states with microbial spatial niches and inter-kingdom signaling (e.g., ligand–receptor and metabolite exchange), providing a foundation for developing new strategies to engineer crop–microbiome interactions to support sustainable agriculture. We conclude by outlining key methodological challenges and future research priorities that point toward building a fully integrated cellular interactome of the plant holobiont. Full article

Background/Objectives: Early childhood caries (ECC) and saliva have been studied across disparate domains, including microbiome, fluoride, immune, oxidative-stress, and neuroendocrine research. However, the ECC–saliva literature has not previously been mapped as a connected system using modern natural language processing (NLP). This study treats PubMed titles and abstracts as data to identify major themes, emerging topics, and candidate neuroimmune axes in ECC–saliva research. Methods: Using the NCBI E-utilities API, we retrieved 298 PubMed records (2000–2025) matching (“early childhood caries” [Title/Abstract]) AND saliva [Title/Abstract]. Text was cleaned with spaCy and embedded using a transformer encoder; BERTopic combined UMAP dimensionality reduction and HDBSCAN clustering to derive thematic topics. We summarised topics with class-based TF–IDF, constructed keyword co-occurrence networks, defined an internal topic-level Novelty Index (semantic distance plus temporal dispersion), and mapped high-novelty topics to gene ontology and Reactome pathways using g:Profiler. Prophet was used to model temporal trends and forecast topic-level publication trajectories. Finally, we generated a fully synthetic neuroimmune salivary dataset, based on realistic ranges from the literature, to illustrate how the identified axes could be operationalised in future ECC cohorts. Results: Seven coherent ECC–saliva topics were identified, including classical microbiome and fluoride domains as well as antioxidant/redox, proteomic, peptide immunity, and Candida–biofilm themes. High-novelty topics clustered around total antioxidant capacity, glutathione peroxidase, superoxide dismutase, and peptide-based host defence. Keyword networks and ontology enrichment highlighted “Detoxification of Reactive Oxygen Species”, “cellular oxidant detoxification”, and cytokine-mediated signalling as central processes. Temporal forecasting suggested plateauing growth for classical epidemiology and fluoride topics, with steeper projected increases for antioxidant and peptide-immunity themes. A co-mention heatmap revealed a literature-level Candida–cytokine–neuroendocrine triad (e.g., Candida albicans, IL-6/TNF, cortisol), which we propose as a testable neuro-immunometabolic hypothesis rather than a confirmed mechanism. Conclusions: AI-assisted topic modelling and network analysis provide a reproducible, bibliometric map of ECC–saliva research that highlights underexplored antioxidant/redox and neuroimmune salivary axes. The synthetic neuroimmune dataset and modelling pipeline are illustrative only, but together with the literature map, they offer a structured agenda for future ECC cohorts and mechanistic studies. Full article

In electroencephalogram (EEG)-based emotion recognition tasks, existing end-to-end approaches predominantly rely on real-valued neural networks, which mainly operate in the time–amplitude domain. However, EEG signals are a type of wave, intrinsically including frequency, phase, and amplitude characteristics. Real-valued architectures may struggle to capture amplitude–phase coupling and spectral structures that are crucial for emotion decoding. To the best of our knowledge, this work is the first to introduce complex-valued neural networks for EEG-based emotion recognition, upon which we design a new end-to-end architecture named Complex-valued EEGNet (CV-EEGNet). Beginning with raw EEG signals, CV-EEGNet transforms them into complex-valued spectra via the Fast Fourier Transform, then sequentially applies complex-valued spectral, spatial, and depthwise-separable convolution modules to extract frequency structures, spatial topologies, and high-level semantic representations while preserving amplitude–phase relationships. Finally, a complex-valued, fully connected classifier generates complex logits, and the final emotion predictions are derived from their magnitudes. Experiments on the SEED (three-class) and SEED-IV (four-class) datasets validate the effectiveness of the proposed method, with t-SNE visualizations further confirming the discriminability of the learned representations. These results show the potential of complex-valued neural networks for raw-signal EEG emotion recognition. Full article

This paper proposes a robust convolutional neural network (CNN) architecture for human activity recognition (HAR) using smartphone accelerometer data, evaluated on the WISDM dataset. We introduce two novel pooling mechanisms—Pooling A (Extrema Contrast Pooling (ECP)) and Pooling B (Center Minus Variation (CMV))—that enhance feature discrimination and noise robustness. ECP emphasizes sharp signal transitions through a nonlinear penalty based on the squared range between extrema, while CMV Pooling penalizes local variability by subtracting the standard deviation, improving resilience to noise. Input data are normalized to the [0, 1] range to ensure bounded and interpretable pooled outputs. The proposed framework is evaluated in two separate configurations: (1) a 1D CNN applied to raw tri-axial sensor streams with the proposed pooling layers, and (2) a histogram-based image encoding pipeline that transforms segment-level sensor redundancy into RGB representations for a 2D CNN with fully connected layers. Ablation studies show that histogram encoding provides the largest improvement, while the combination of ECP and CMV further enhances classification performance. Across six activity classes, the 2D CNN system achieves up to 96.84% weighted classification accuracy, outperforming baseline models and traditional average pooling. Under Gaussian, salt-and-pepper, and mixed noise conditions, the proposed pooling layers consistently reduce performance degradation, demonstrating improved stability in real-world sensing environments. These results highlight the benefits of redundancy-aware pooling and histogram-based representations for accurate and robust mobile HAR systems. Full article

This work investigates the effectiveness of block transform coding (BTC) as a lightweight, training-free quantization strategy for compressing the weights of pretrained deep neural networks. The proposed method applies a rule-based block transform with variance and root mean square error (RMSE)-driven stopping criteria, enabling substantial reductions in bit precision while preserving the statistical structure of convolutional and fully connected layer weights. Unlike uniform 8-bit quantization, BTC dynamically adjusts bit usage across layers and achieves significantly lower distortion for the same compression budget. We evaluate BTC across many pretrained architectures and tabular benchmarks. Experimental results show that BTC consistently reduces storage to 4–7.7 bits per weight while maintaining accuracy within 2–3% of the 32-bit floating point (FP32) baseline. To further assess scalability and baseline strength, BTC is additionally evaluated on large-scale ImageNet models and compared against a calibrated percentile-based uniform post-training quantization method. The results show that BTC achieves a substantially lower effective bit-width while incurring only a modest accuracy reduction relative to calibration-aware 8-bit quantization, highlighting a favorable compression–accuracy trade-off. BTC also exhibits stable behavior across successive post-training quantization (PTQ) configurations, low quantization noise, and smooth RMSE trends, outperforming naïve uniform quantization under aggressive compression. These findings confirm that BTC provides a scalable, architecture-agnostic, and training-free quantization mechanism suitable for deployment in memory- and computing-constrained environments. Full article

The optimization of low-Reynolds-number airfoils for micro unmanned aerial vehicles (UAVs) is challenging due to strong geometric nonlinearities, tight endurance requirements, and the need to maintain performance across multiple operating conditions. Classical surrogate-assisted optimization (SAO) methods combined with genetic algorithms become increasingly expensive and less reliable when class–shape transformation (CST)-based geometries are coupled with several flight conditions. Although deep learning surrogates have higher expressive power, their use in this context is often limited by insufficient local feature extraction, weak adaptation to changes in operating conditions, and a lack of robustness analysis. In this study, we construct a task-specific convolutional neural network–long short-term memory (CNN–LSTM) surrogate that jointly predicts the power factor, lift, and drag coefficients at three representative operating conditions (cruise, forward flight, and maneuver) for the same CST-parameterized airfoil and integrate it into an Non-dominated Sorting Genetic Algorithm II (NSGA-II)-based three-objective optimization framework. The CNN encoder captures local geometric sensitivities, while the LSTM aggregates dependencies across operating conditions, forming a compact encoder–aggregator tailored to low-Re micro-UAV design. Trained on a computational fluid dynamics (CFD) dataset from a validated SD7032-based pipeline, the proposed surrogate achieves substantially lower prediction errors than several fully connected and single-condition baselines and maintains more favorable error distributions on CST-family parameter-range extrapolation samples (±40%, geometry-valid) under the same CFD setup, while being about three orders of magnitude faster than conventional CFD during inference. When embedded in NSGA-II under thickness and pitching-moment constraints, the surrogate enables efficient exploration of the design space and yields an optimized airfoil that simultaneously improves power factor, reduces drag, and increases lift compared with the baseline SD7032. This work therefore contributes a three-condition surrogate–optimizer workflow and physically interpretable low-Re micro-UAV design insights, rather than introducing a new generic learning or optimization algorithm. Full article

Beamforming is highly significant for the physical layer of wireless communication systems, for multi-antenna systems such as multiple input multiple output (MIMO) and massive MIMO, since it improves spectral efficiency and reduces interference. Traditional linear beamforming methods such as zero-forcing beamforming (ZFBF) and minimum mean square error (MMSE) beamforming provide closed-form solutions. Yet, their performance drops when they face non-ideal conditions such as imperfect channel state information (CSI), dynamic propagation environment, or high-dimensional system configurations, primarily due to static assumptions and computational limitations. These limitations have led to the rise of deep learning-based beamforming, where data-driven models derive beamforming solutions directly from CSI. By leveraging the representational capabilities of cutting-edge deep learning architectures, along with the increasing availability of data and computational resources, deep learning presents an adaptive and potentially scalable alternative to traditional methodologies. In this work, we unify and systematically compare our two unsupervised learning architectures for uplink receive beamforming: a simple neural network beamforming (NNBF) model, composed of convolutional and fully connected layers, and a transformer-based NNBF model that integrates grouped convolutions for feature extraction and transformer blocks to capture long-range channel dependencies. They are evaluated in a common multi-user single input multiple output (MU-SIMO) system model to maximize sum-rate across single-antenna user equipments (UEs) under 3GPP-compliant channel models, namely TDL-A and UMa. Furthermore, we present a FLOPs-based asymptotic computational complexity analysis for the NNBF architectures alongside baseline methods, namely ZFBF and MMSE beamforming, explicitly characterizing inference-time scaling behavior. Experiments for the simple NNBF are performed under simplified assumptions such as stationary UEs and perfect CSI across varying antenna configurations in the TDL-A channel. On the other hand, transformer-based NNBF is evaluated in more realistic conditions, including urban macro environments with imperfect CSI, diverse UE mobilities, coding rates, and modulation schemes. Results show that the transformer-based NNBF achieves superior performance under realistic conditions at the cost of increased computational complexity, while the simple NNBF presents comparable or better performance than baseline methods with significantly lower complexity under simplified assumptions. Full article

The rapid growth of dockless electric scooter (e-scooter) sharing services has transformed short-distance urban mobility, offering convenience and sustainability benefits while amplifying challenges related to demand imbalance, fleet rebalancing, and spatial inequity. Accurate spatiotemporal demand prediction is therefore essential for optimizing resource allocation and supporting data-driven policy interventions. This study proposes a hybrid deep learning framework that integrates a Graph Convolutional Network (GCN) with a Gated Recurrent Unit (GRU) and community detection to enhance short-term prediction of e-scooter pick-up and drop-off demands. The Louvain algorithm is employed to partition urban areas into mobility-based communities, enabling the model to capture functional connectivity rather than relying solely on geographic proximity. Using real-world e-scooter trip data from Calgary, Canada, the model’s performance is evaluated against established baselines, including a Masked Fully Convolutional Network (MFCN) and conventional GRU architectures. Results show that the proposed approach achieves up to 11.8% improvement in mean absolute error (MAE) compared with the MFCN baseline and more robust generalization across temporal horizons. The findings demonstrate that integrating community structures into graph-based learning effectively captures complex urban dynamics, providing practical insights for sustainable micromobility operation and service deployment. Full article

The rise of connected and automated vehicles has transformed in-vehicle infotainment (IVI) systems into critical gateways linking user interfaces, vehicular networks, and cloud-based fleet services. A concerning architectural reality is that hardcoded credentials like access point names (APNs) in IVI firmware create a cross-layer attack surface where local exposure can escalate into entire vehicle fleets being remotely compromised. To address this risk, we propose a cross-layer security framework that integrates firmware extraction, symbolic execution, and targeted fuzzing to reconstruct authentic IVI-to-backend interactions and uncover high-impact web vulnerabilities such as server-side request forgery (SSRF) and broken access control. Applied across seven diverse automotive systems, including major original equipment manufacturers (OEMs) (Mercedes-Benz, Tesla, SAIC, FAW-VW, Denza), Tier-1 supplier Bosch, and advanced driver assistance systems (ADAS) vendor Minieye, our approach exposes systemic anti-patterns and demonstrates a fully realized exploit that enables remote control of approximately six million Mercedes-Benz vehicles. All 23 discovered vulnerabilities, including seven CVEs, were patched within one month. In closed automotive ecosystems, we argue that the true measure of efficacy lies not in maximizing code coverage but in discovering actionable, fleet-wide attack paths, which is precisely what our approach delivers. Full article