Search Results (1,817)

Efficient metadata and page management are essential for sustaining database performance on modern flash-based storage. However, conventional SQLite configurations—rollback journal and WAL—often trigger excessive small writes and frequent synchronization events, leading to high write amplification and degraded tail latency, particularly on UFS and NVMe devices. This study introduces ALEX (Adaptive Log-Embedded Extent Layer), a lightweight VFS-level extension that coalesces scattered 4 KB page updates into sequential, page-aligned extents while embedding compact log records for recovery. The proposed design reduces redundant writes through in-memory page deduplication, minimizes fdatasync()frequency by flushing multi-page extents, and preserves full SQLite compatibility. We evaluate ALEX on both Linux NVMe SSDs and Android UFS storage under controlled workloads. Results show that ALEX significantly lowers write amplification, reduces sync counts, and improves p95–p99 write latency compared with baseline SQLite modes. The approach consistently achieves near-sequential write patterns without modifying SQLite internals. These findings demonstrate that lightweight extent-based coalescing can provide substantial efficiency gains for embedded and mobile database systems, offering a practical direction for enhancing SQLite performance on flash devices. Full article

Symmetry and asymmetry are foundational design principles in artificial intelligence, defining the balance between invariance and adaptability in multimodal learning systems. In audio-visual speaker verification, where speech and lip-motion features are jointly modeled to determine whether two utterances belong to the same individual, these principles govern both fairness and discriminative power. In this work, we analyze how symmetry and asymmetry emerge within a gated-fusion architecture that integrates Time-Delay Neural Networks and Bidirectional Long Short-Term Memory encoders for speech, ResNet-based visual lip encoders, and a shared Conformer-based temporal backbone. Structural symmetry is preserved through weight-sharing across paired utterances and symmetric cosine-based scoring, ensuring verification consistency regardless of input order. In contrast, asymmetry is intentionally introduced through modality-dependent temporal encoding, multi-head attention pooling, and a learnable gating mechanism that dynamically re-weights the contribution of audio and visual streams at each timestep. This controlled asymmetry allows the model to rely on visual cues when speech is noisy, and conversely on speech when lip visibility is degraded, yielding adaptive robustness under cross-modal degradation. Experimental results demonstrate that combining symmetric embedding space design with adaptive asymmetric fusion significantly improves generalization, reducing Equal Error Rate (EER) to 3.419% on VoxCeleb-2 test dataset without sacrificing interpretability. The findings show that symmetry ensures stable and fair decision-making, while learnable asymmetry enables modality awareness together forming a principled foundation for next-generation audio-visual speaker verification systems. Full article

The paper focuses on automated diagnosis of retinal diseases, particularly Age-related Macular Degeneration (AMD) and diabetic retinopathy (DR), using optical coherence tomography (OCT), while addressing three key challenges: disease comorbidity, severe class imbalance, and the lack of strictly paired OCT and fundus data. We propose a hierarchical modular deep learning system designed for multi-label OCT screening with conditional routing to specialized staging modules. To enable DR staging when fundus images are unavailable, we use cross-modal alignment between OCT and fundus representations. This approach involves training a latent bridge that projects OCT embeddings into the fundus feature space. We enhance clinical reliability through per-class threshold calibration and implement quality control checks for OCT-only DR staging. Experiments demonstrate robust multi-label performance (macro-F1 $=0.989\pm 0.006$ after per-class threshold calibration) and reliable calibration (ECE $=2.1\pm 0.4\%$), and OCT-only DR staging is feasible in 96.1% of cases that meet the quality control criterion. Full article

Reliable data availability and transparent governance are fundamental requirements for distributed edge-to-cloud systems that must operate across multiple administrative domains. Conventional cloud-centric architectures centralize control and storage, creating bottlenecks and limiting autonomous collaboration at the network edge. This paper introduces a decentralized governance and service-management framework that leverages Decentralized Autonomous Organizations (DAOs) and Decentralized Applications (DApps) to to govern and orchestrate verifiable, tamper-resistant, and continuously accessible data exchange between heterogeneous edge and cloud components. By embedding blockchain-based smart contracts within swarm-enabled edge infrastructures, the approach enables automated decision-making, auditable coordination, and fault-tolerant data sharing without relying on trusted intermediaries. The proposed OASEES framework demonstrates how DAO-driven orchestration can enhance data availability and accountability in real-world scenarios, including energy grid balancing, structural safety monitoring, and predictive maintenance of wind turbines. Results highlight that decentralized governance mechanisms enhance transparency, resilience, and trust, offering a scalable foundation for next-generation edge-to-cloud data ecosystems. Full article

Bridge health monitoring is crucial for ensuring the safety and durability of infrastructure. In traditional methods, physics-based models have high interpretability but are difficult to handle complex nonlinear problems, while purely data-driven machine learning methods are limited by data scarcity and physical inconsistency. Physics-informed machine learning, as an emerging “gray box” paradigm, effectively integrates the advantages of both by embedding physical laws (such as control equations) into machine learning models in the form of constraints, priors, or residuals. This article systematically elaborates on the core fusion mechanism of physics-informed machine learning (PIML) in bridge engineering, innovative applications throughout the entire lifecycle of design, construction, operation, and maintenance, as well as its unique data augmentation strategy. Research has shown that PIML can significantly improve the accuracy and robustness of damage identification, load inversion, and performance prediction, and is the core engine for constructing dynamic and predictive digital twin systems. Despite facing challenges in complex physical modeling, loss function balancing, and engineering interpretability, PIML represents a fundamental shift in bridge health monitoring towards intelligent and predictive maintenance by combining advanced strategies such as active learning and meta learning with IoT technology. Full article

Background: Early diagnosis of Alzheimer’s disease (AD) is essential for slowing disease progression and mitigating cognitive decline. However, conventional diagnostic methods are often invasive, time-consuming, and costly, limiting their utility in large-scale screening. There is an urgent need for scalable, non-invasive, and accessible screening tools. Methods: We propose a novel screening framework combining a pre-trained multimodal large language model with structured MMSE speech tasks. An artificial intelligence-assisted multilingual Mini-Mental State Examination system (AAM-MMSE) was utilized to collect voice data from 1098 participants in Sichuan and Chongqing. CosyVoice2 was used to extract speaker embeddings, speech labels, and acoustic features, which were converted into statistical representations. Fourteen machine learning models were developed for subject classification into three diagnostic categories: Healthy Control (HC), Mild Cognitive Impairment (MCI), and Alzheimer’s Disease (AD). SHAP analysis was employed to assess the importance of the extracted speech features. Results: Among the evaluated models, LightGBM and Gradient Boosting classifiers exhibited the highest performance, achieving an average AUC of 0.9501 across classification tasks. SHAP-based analysis revealed that spectral complexity, energy dynamics, and temporal features were the most influential in distinguishing cognitive states, aligning with known speech impairments in early-stage AD. Conclusions: This framework offers a non-invasive, interpretable, and scalable solution for cognitive screening. It is suitable for both clinical and telemedicine applications, demonstrating the potential of speech-based AI models in early AD detection. Full article

Upper-limb motor impairment is a major consequence of stroke and neuromuscular disorders, imposing a sustained clinical and socioeconomic burden worldwide. Quantitative assessment of limb positioning and motion accuracy is fundamental to rehabilitation, guiding therapy evaluation and robotic assistance. The evolution of upper-limb positioning systems has progressed from optical motion capture to wearable inertial measurement units (IMUs) and, more recently, to data-driven estimators integrated with rehabilitation robots. Each generation has aimed to balance spatial accuracy, portability, latency, and metrological reliability under ecological conditions. This review presents a systematic synthesis of the state of measurement uncertainty, calibration, and traceability in upper-limb rehabilitation robotics. Studies are categorised across four layers, i.e., sensing, fusion, cognitive, and metrological, according to their role in data acquisition, estimation, adaptation, and verification. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) protocol was followed to ensure transparent identification, screening, and inclusion of relevant works. Comparative evaluation highlights how modern sensor-fusion and learning-based pipelines achieve near-optical angular accuracy while maintaining clinical usability. Persistent challenges include non-standard calibration procedures, magnetometer vulnerability, limited uncertainty propagation, and absence of unified traceability frameworks. The synthesis indicates a gradual transition toward cognitive and uncertainty-aware rehabilitation robotics in which metrology, artificial intelligence, and control co-evolve. Traceable measurement chains, explainable estimators, and energy-efficient embedded deployment emerge as essential prerequisites for regulatory and clinical translation. The review concludes that future upper-limb systems must integrate calibration transparency, quantified uncertainty, and interpretable learning to enable reproducible, patient-centred rehabilitation by 2030. Full article

Fault-tolerant control in safety-critical industrial systems demands adaptive responses to equipment degradation, parameter drift, and sensor failures while maintaining strict operational constraints. Traditional model-based controllers struggle under these conditions, requiring extensive retuning and dense instrumentation. Recent safe multi-agent reinforcement learning (MARL) frameworks with control barrier functions (CBFs) achieve real-time constraint satisfaction in robotics and power systems, yet assume comprehensive state observability—incompatible with sensor-hostile industrial environments where instrumentation degradation and contamination risks dominate design constraints. This work presents a safety-aware multi-agent deep reinforcement learning framework for adaptive fault-tolerant control in sensor-lean industrial environments, achieving formal safety through learned implicit barriers under partial observability. The framework integrates four synergistic mechanisms: (1) multi-layer safety architecture combining constrained action projection, prioritized experience replay, conservative training margins, and curriculum-embedded verification achieving zero constraint violations; (2) multi-agent coordination via decentralized execution with learned complementary policies. Additional components include (3) curriculum-driven sim-to-real transfer through progressive four-stage learning achieving 85–92% performance retention without fine-tuning; (4) offline extended Kalman filter validation enabling 70% instrumentation reduction (91–96% reconstruction accuracy) for regulatory auditing without real-time estimation dependencies. Validated through sustained deployment in commercial beverage manufacturing clean-in-place (CIP) systems—a representative safety-critical testbed with hard flow constraints (≥1.5 L/s), harsh chemical environments, and zero-tolerance contamination requirements—the framework demonstrates superior control precision (coefficient of variation: 2.9–5.3% versus 10% industrial standard) across three hydraulic configurations spanning complexity range 2.1–8.2/10. Comprehensive validation comprising 37+ controlled stress-test campaigns and hundreds of production cycles (accumulated over 6 months) confirms zero safety violations, high reproducibility (CV variation < 0.3% across replicates), predictable complexity–performance scaling ($R^2=0.89$), and zero-retuning cross-topology transferability. The system has operated autonomously in active production for over 6 months, establishing reproducible methodology for safe MARL deployment in partially-observable, sensor-hostile manufacturing environments where analytical CBF approaches are structurally infeasible. Full article

Polymer-assisted hot water flooding (PAHWF) is an important enhanced oil recovery technique involving strongly coupled thermal, chemical, and multiphase flow processes. Accurate prediction of water saturation, polymer concentration, and temperature evolution in PAHWF is challenging due to the highly nonlinear and multiscale governing equations. In this study, a physics-informed neural network (PINN) framework is developed for one-dimensional PAHWF simulation as a controlled benchmark system to systematically investigate PINN behavior in multiphysics-coupled problems. The PAHWF governing equations incorporating temperature- and concentration-dependent viscosity are embedded into the PINN loss function. Three progressively designed numerical examples are conducted to examine the effects of temperature normalization, network architecture (PINN-1 versus PINN-2), and network depth on training stability and solution accuracy. The results demonstrate that temperature normalization effectively mitigates gradient-scale imbalance, significantly improving convergence stability and prediction accuracy. Furthermore, the PINN-2 architecture, which employs a dedicated network for temperature, exhibits enhanced robustness and accuracy compared with the unified PINN-1 structure. Variations in network depth show limited influence on solution quality, indicating the inherent robustness of PINNs under the proposed framework. Although conventional numerical methods remain more efficient for one-dimensional forward problems, this study establishes a methodological foundation for extending PINNs to higher-dimensional, strongly coupled PAHWF simulations and inverse reservoir problems. The proposed framework provides insights into improving PINN trainability and reliability for complex enhanced oil recovery processes. Full article

Background/Objectives: Home care plays a crucial role in contemporary healthcare systems, particularly in the long-term care of people with chronic and progressive illnesses. Family caregivers often experience substantial physical, emotional, and organizational burden. Telemedicine and digital health applications have the potential to support home care by improving health monitoring, communication, and care coordination. However, their use among family caregivers remains inconsistent, and little is known about how organizational support structures such as telemedicine centers influence acceptance and everyday use. This study aims to examine the benefits of telemedicine in home care and to evaluate the role of telemedicine centers as supportive infrastructures for family caregivers. Methods: A mixed-methods design was applied. Quantitative data were collected through an online survey of 58 family caregivers to assess the use of telemedicine and digital health applications, perceived benefits, barriers, and support needs. This was complemented by an in-depth qualitative case study exploring everyday caregiving experiences with telemedicine technologies and telemedicine center support. A systematic literature review informed the theoretical framework and the development of the empirical instruments. Results: Most respondents reported not using telemedicine or digital health applications in home care. Among users, telemedicine was associated with perceived improvements in quality of care, particularly through enhanced health monitoring, improved communication with healthcare professionals, and increased feelings of safety and control. Key barriers to adoption included technical complexity, data protection concerns, and limited digital literacy. Both quantitative findings and the qualitative case study highlighted the importance of structured support. Telemedicine centers were perceived as highly beneficial, providing technical assistance, training, coordination, and ongoing guidance that facilitated technology acceptance and sustained use. Conclusions: Telemedicine and digital health applications can meaningfully support home care and reduce caregiver burden when they are embedded in supportive socio-technical structures. Telemedicine centers can function as central points of contact that enhance usability, trust, and continuity of care. The findings suggest that successful implementation of telemedicine in home care requires not only technological solutions but also accessible organizational support and targeted training for family caregivers. Full article

This work advances a 6G-ready, micro-granular SDN fabric that unifies high-performance edge data planes with intent-driven, multi-domain orchestration and cloud offloading. First, edge and cell-site whiteboxes are upgraded with Smart Network Interface Cards and embedded AI accelerators, enabling line-rate processing of data flows and on-box learning/inference directly in the data plane. This pushes functions such as traffic classification, telemetry, and anomaly mitigation to the point of ingress, reducing latency and backhaul load. Second, an SDN controller, i.e., ETSI TeraFlowSDN, is extended to deliver multi-domain SDN orchestration with native lifecycle management (LCM) for whitebox Network Operating Systems—covering onboarding, configuration-drift control, rolling upgrades/rollbacks, and policy-guarded compliance—so operators can reliably manage heterogeneous edge fleets at scale. Third, the SDN controller incorporates a new NFV-O client that seamlessly offloads network services—such as ML pipelines or NOS components—to telco clouds via an NFV orchestrator (e.g., ETSI Open Source MANO), enabling elastic placement and scale-out across the edge–cloud continuum. Together, these contributions deliver an open, programmable platform that couples in-situ acceleration with closed-loop, intent-based orchestration and elastic cloud resources, targeting demonstrable gains in end-to-end latency, throughput, operational agility, and energy efficiency for emerging 6G services. Full article

Unmanned Aerial Spraying Systems (UASS) has rapidly advanced precision crop protection. However, the spray performance of UASSs is influenced by nozzle atomization, rotor-induced airflow, and external environmental conditions. These factors cause strong spatiotemporal coupling and high uncertainty. As a result, visualization-based monitoring techniques are now essential for understanding these dynamics and supporting spray modeling and drift-mitigation design. This review highlights developments in spray visualization technologies along the “droplet–airflow–target” chain mechanism in UASS spraying. We first outline the physical fundamentals of droplet formation, liquid-sheet breakup, droplet size distribution, and transport mechanisms in rotor-induced flow. Dominant processes are identified across near-field, mid-field, and far-field scales. Next, we summarize major visualization methods. These include optical imaging (PDPA/PDIA, HSI, DIH), laser-based scattering and ranging (LD, LiDAR), and flow-field visualization (PIV). We compare their spatial resolution, measurement range, 3D reconstruction capabilities, and possible sources of error. We then review wind-tunnel trials, field experiments, and point-cloud reconstruction studies. These studies show how downwash flow and tip vortices affect plume structure, canopy disturbance, and deposition patterns. Finally, we discuss emerging intelligent analysis for large-scale monitoring—such as image-based droplet recognition, multimodal data fusion, and data-driven modeling. We outline future directions, including unified feature systems, vortex-coupled models, and embedded closed-loop spray control. This review is a comprehensive reference for advancing UASS analysis, drift assessment, spray optimization, and smart support systems. Full article

Road transport systems are central to sustainable mobility and the energy transition because they account for a large share of final energy use and remain heavily dependent on fossil fuels. With more than 90% of transport energy still supplied by petroleum-based fuels, improving energy efficiency and reducing emissions in road networks has become a strategic priority. This review compares Australia, Hong Kong, and the United Kingdom to examine how road-design standards and emerging digital technologies can improve energy performance across planning, design, operations, and maintenance. Using Australia’s Austroads Guide to Road Design, Hong Kong’s Transport Planning and Design Manual (TPDM), and the UK’s Design Manual for Roads and Bridges (DMRB) as core reference frameworks, we apply a rubric-based document analysis that codes provisions by mechanism type (direct, indirect, or emergent), life-cycle stage, and energy relevance. The findings show that energy-relevant outcomes are embedded through different pathways: TPDM most strongly supports urban operational efficiency via coordinated/adaptive signal control and public-transport prioritization; DMRB emphasizes strategic-network flow stability and whole-life carbon governance through managed motorway operations and life-cycle assessment requirements; and Austroads provides context-sensitive, performance-based guidance that supports smoother operations and active travel, with implementation varying by jurisdiction. Building on these results, the paper proposes an AI-enabled benchmarking overlay that links manual provisions to comparable energy and carbon indicators to support cross-jurisdictional learning, investment prioritization, and future manual revisions toward safer, more efficient, and low-carbon road transport systems. Full article

Background/Objectives: Cardiovascular (CV) diseases remain the leading cause of morbidity and mortality across Europe. Despite substantial progress in prevention, diagnostics, and therapeutics, outpatient cardiology care continues to face systemic challenges, including limited consultation time, workforce constraints, and incomplete clinical information at the point of care. The primary objective of this study is threefold. First, to evaluate whether AI-enhanced telemedicine improves clinical control of hypertension, dyslipidemia, and heart failure compared to standard ambulatory care. Second, to assess the impact on physician workflow efficiency and documentation burden through AI-assisted clinical documentation. Third, to determine patient satisfaction and safety profiles of integrated telemedicine–AI systems. Clinical control will be measured by a composite endpoint of disease-specific targets assessed at the 12-month follow-up visit. Methods: The TeleAI-CVD Concept Study aims to evaluate the integration of telemedicine and artificial intelligence (AI) to enhance the efficiency, quality, and individualization of cardiovascular disease management in the ambulatory setting. Within this framework, AI-driven tools will be employed to collect structured clinical histories and current symptomatology from patients prior to outpatient visits using digital questionnaires and conversational interfaces. Results: Obtained data, combined with telemonitoring metrics, laboratory parameters, and existing clinical records, will be synthesized to support clinical decision-making. Conclusions: This approach is expected to streamline consultations, increase diagnostic accuracy, and enable personalized, data-driven care through continuous evaluation of patient trajectories. The anticipated outcomes of the TeleAI-CVD study include the development of optimized, AI-assisted management protocols for cardiology patients, a reduction in unnecessary in-person visits through effective telemedicine-based follow-up, and accelerated attainment of therapeutic targets. Ultimately, this concept seeks to redefine the paradigm of outpatient cardiovascular care by embedding advanced digital technologies within routine clinical workflows. Full article

Background/Objectives: Glioblastoma (GBM) exhibits heterogeneous, nonlinear invasion patterns that challenge conventional modeling and radiomic prediction. Most deep learning approaches describe the morphology but rarely capture the dynamical stability of tumor evolution. We propose an AI framework that approximates a latent attractor landscape of GBM morphodynamics—stable basins in a continuous manifold that are consistent with reproducible morphologic regimes. Methods: Multimodal MRI scans from BraTS 2020 (n = 494) were standardized and embedded with a 3D autoencoder to obtain 128-D latent representations. Unsupervised clustering identified latent basins (“attractors”). A neural ordinary differential equation (neural-ODE) approximated latent dynamics. All dynamics were inferred from cross-sectional population variability rather than longitudinal follow-up, serving as a proof-of-concept approximation of morphologic continuity. Voxel-level perturbation quantified local morphodynamic sensitivity, and proof-of-concept control was explored by adding small inputs to the neural-ODE using both a deterministic controller and a reinforcement learning agent based on soft actor–critic (SAC). Survival analyses (Kaplan–Meier, log-rank, ridge-regularized Cox) assessed associations with outcomes. Results: The learned latent manifold was smooth and clinically organized. Three dominant attractor basins were identified with significant survival stratification (χ² = 31.8, p = 1.3 × 10⁻⁷) in the static model. Dynamic attractor basins derived from neural-ODE endpoints showed modest and non-significant survival differences, confirming that these dynamic labels primarily encode the morphodynamic structure rather than fixed prognostic strata. Dynamic basins inferred from neural-ODE flows were not independently prognostic, indicating that the inferred morphodynamic field captures geometric organization rather than additional clinical risk information. The latent stability index showed a weak but borderline significant negative association with survival (ρ = −0.13 [−0.26, −0.01]; p = 0.0499). In multivariable Cox models, age remained the dominant covariate (HR = 1.30 [1.16–1.45]; p = 5 × 10⁻⁶), with overall C-indices of 0.61–0.64. Voxel-level sensitivity maps highlighted enhancing rims and peri-necrotic interfaces as influential regions. In simulation, deterministic control redirected trajectories toward lower-risk basins (≈57% success; ≈96% terminal distance reduction), while a soft actor–critic (SAC) agent produced smoother trajectories and modest additional reductions in terminal distance, albeit without matching the deterministic controller’s success rate. The learned attractor classes were internally consistent and clinically distinct. Conclusions: Learning a latent attractor landscape links generative AI, dynamical systems theory, and clinical outcomes in GBM. Although limited by the cross-sectional nature of BraTS and modest prognostic gains beyond age, these results provide a mechanistic, controllable framework for tumor morphology in which inferred dynamic attractor-like flows describe latent organization rather than a clinically predictive temporal model, motivating prospective radiogenomic validation and adaptive therapy studies. Full article