Search Results (3,643)

This study presents a data-driven framework for the mechanistic interpretation of asphalt pavement responses using an integrated smart sensing and monitoring system deployed on a national highway in Thailand. A fully instrumented pavement test section was developed, incorporating a multi-sensor embedded network and a field data acquisition platform integrated with weigh-in-motion (WIM) technology. The system consists of 54 sensors, including strain gauges, pressure cells, moisture sensors, and thermocouples, installed at multiple depths to capture high-resolution stress–strain responses under controlled heavy-vehicle loading. Field measurements were analyzed and compared with classical mechanistic models, including Boussinesq’s theory, Odemark’s equivalent thickness method, and Burmister’s multilayer elastic theory. The results demonstrate good agreement for vertical stress predictions in deeper layers, while significant discrepancies were observed in strain responses, particularly in the asphalt layer, where measured tensile strains were up to 2.5 times higher than theoretical estimates. The findings indicate that conventional elastic models provide useful first-order approximations; however, discrepancies were observed in representing the viscoelastic behavior of asphalt materials under real loading conditions. Furthermore, the integration of sensor data with traffic loading information confirms that axle load magnitude is the dominant factor governing pavement responses, whereas vehicle speed primarily influences load duration. The proposed framework demonstrates the potential of smart sensing systems for enabling automated, data-driven pavement analysis and supporting digital twin-based infrastructure management. Full article

Post-processing remains a major bottleneck in the fabrication of microfluidic devices using Digital Light Processing Stereolithography (DLP-SLA) 3D printing, where unpolymerized resin trapped within internal structures must be removed without damaging delicate features such as thin membranes, valves, and pumps. Manual flushing is slow, inconsistent, and prone to structural failure, especially as device complexity and port counts increase. Here, we present the first fully automated flushing system for DLP-SLA microfluidic devices, enabled by a standardized chip-to-chip (C2C) interconnect architecture and an electronically controlled pneumatic routing platform. A reusable 32-port flushing interface chip provides alignment, sealing, and modular coupling to arbitrary device chips through integrated microgaskets, while a network of electronic pressure controllers, differential pressure sensors, and multi-port rotary valves enable precise, programmable application of pressure, vacuum, and solvent conditions. We introduce a fluidic-circuit model of the system that relates applied pressure to the pressure drop across device structures and experimentally validate this model using channels with varying fluidic resistances. Using this platform, we demonstrate robust flushing of both passive (straight and serpentine channels) and active (valves, pumps) microfluidic elements, as well as application-specific devices including mixers and concentration-gradient generators. Our system eliminates manual handling, improves valve membrane survival, and provides repeatable flushing across a broad range of device geometries. This work establishes a scalable foundation for automated post-processing in 3D-printed microfluidics and significantly advances the practicality of DLP-SLA fabrication for complex, multi-layered microfluidic devices. Full article

Explainable Artificial Intelligence (XAI) has become a critical requirement in medical image analysis, where transparency and interpretability are essential for clinical trust and decision support. Melanoma is recognized as one of the most deadly types of skin cancer, with its occurrence exhibiting an increasing pattern in recent times. However, detecting this cancer in its initial stages greatly increases patients’ chances of long-term survival. Various computer-based techniques have recently been proposed to diagnose skin lesions at their early stages. Even though the machine learning community has achieved a certain degree of success, there is still an unresolved research challenge regarding high error margins and the limited interpretability of automated systems. This study focuses on addressing both segmentation and classification tasks, with particular emphasis on two key concepts: (1) improving image quality to maximize distinguishability between foreground and background regions, thereby enhancing visual interpretability and segmentation accuracy and (2) eliminating redundant and cluttered feature information to generate the most discriminative and compact feature representations. The input images are initially processed using a novel metaheuristic contrast-stretching method to estimate image-specific key parameters, thereby enhancing lesion boundary clarity in a clinically interpretable manner. Following this, the improved images are fed into selected pre-trained deep models, including DenseNet-201, Inception-ResNet v2, and NASNet-Mobile. The extracted features from all pre-trained models are fused to produce resultant vectors, which are then refined using a bio-inspired feature selection method, termed entropy-controlled whale optimization, to retain only the most informative attributes. The selected discriminative feature set is subsequently classified using multiple classifiers. The results indicate that the proposed framework achieves superior performance compared to existing methods in terms of accuracy, sensitivity, specificity, and F1-score. Additionally, it facilitates a more explainable, transparent, and structured diagnostic pipeline appropriate for medical applications. Full article

Construction monitoring systems can benefit from automated detection of both structural components and personal protective equipment (PPE). While previous studies focus on single-task applications using YOLOv5-YOLOv10 architectures, this study presents a systematic comparative evaluation of four neck-level architectural modifications within the YOLOv11n framework: Depthwise separable convolutions (DSC) for computational efficiency, multi-scale dilated convolutions (MSDC) for expanded receptive fields, feature refinement interfaces (FRI) for learnable feature adaptation, and dual attention mechanisms (DAM) for enhanced feature discriminability. Controlled experiments were conducted on precast components (3771 images, 6 classes) and PPE (5201 images, 3 classes) datasets. DAM-YOLO achieved the highest performance with mAP@50 of 0.972 (precast) and 0.968 (PPE), while the performance range across all variants spanned 0.942–0.972 (precast) and 0.936–0.968 (PPE). All variants demonstrated robust detection capabilities with mAP@50 ≥ 0.936 and mAP@50–95 spanning 0.760–0.825. The narrow performance differences across variants indicate that different neck-level modifications yield comparable detection accuracy, providing empirical evidence to support architecture selection within these two evaluated construction object detection scenarios. Full article

Despite the productivity and economic limitations imposed by environmental and climatic conditions, livestock systems play a fundamental role in preserving habitats and high-conservation-value species, while delivering a broad spectrum of ecosystem services to rural populations. Breeders need timely information to produce safe, inexpensive, environmentally, and welfare-friendly food products. Information on feeding and nutrition is of particular importance since it represents a significant percentage of animal breeding costs. Automating the collection, analysis, and use of production-related information on livestock feeding systems represents one of the central challenges facing the sector. Precision feeding systems (PFSs) have deeply changed farm management by providing new information on the health status of animals, their welfare, and nutritional requirements. PFSs encompass modern electronic and ICT-related (information and communication technologies) technologies that facilitate the electronic measurement of critical components, ensuring optimum efficiency of both resource use and animal productivity. This review analyzes the current state and potential applications of precision feeding systems for sustainable livestock production. The implementation and feasibility of PFSs have been investigated across the major animal production species and contexts. Based on the available literature, real-time monitoring and control systems can improve the production efficiency of livestock farms. However, further research is needed, as several components of PFSs are still at different stages of development and commercial readiness. Full article

Automated multiplex immunohistochemistry (IHC) and in situ hybridization (ISH) require staining platforms that combine stable reagent exchange, low-volume operation, process observability, and protocol flexibility. Existing autostainers are often rigid and costly, whereas microfluidic and sensing solutions remain largely component-specific rather than system-oriented. This study proposes and partially validates a layered cyber-physical architecture for multiplex histological staining. The architecture integrates five functional layers—biochemical workflow, fluidic processing, capacitive sensing, protocol-driven control, and software-based process representation—within a unified formal framework and is supported at the subsystem level by experimental characterization of its fluidic and sensing layers. Fluidic experiments on a slot-type microfluidic chamber identified a practical operating window in which upper-feed operation, moderate calibrated flow conditions, and low chamber angles between 10° and 40° provide stable filling and acceptable drainage. The differential slot-line capacitive sensing subsystem detected liquid volumes as low as 0.5 µL, with stable threshold-based interpretation at a practical detection threshold of approximately 5 fF after digital filtering. The control and software layers are specified at the architectural and formal model level; their hardware implementation and closed-loop validation remain subjects of future work. Together, the reported results demonstrate that controlled reagent transport and sensing-based process observability are jointly feasible within the proposed modular framework, establishing a conceptual and experimental foundation for scalable, flexible, and resource-efficient multiplex IHC/ISH systems. Full article

This paper presents a modular system architecture and an automated signal processing pipeline designed to quantify neurophysiological relaxation responses to fragrance using consumer-grade wearable electroencephalography (EEG). By integrating real-time data streaming via Open Sound Control (OSC) with a high-performance backend, the platform enables objective assessment of olfactory stimuli through a reproducible Sleep Readiness Index (SRI) derived from spectral power shifts. To mitigate the signal quality constraints inherent in portable hardware, the framework utilizes a robust suite of engineering controls, including zero-phase filtering and automated artifact rejection, ensuring data integrity across short-window trials. Validation through construct-level analysis of public sleep datasets and synthetic sensitivity testing confirms the index’s directional reliability, while runtime benchmarking demonstrates sub-millisecond compute times suitable for interactive wellness applications. Ultimately, this framework provides a transparent, auditable engineering scaffold that replaces subjective self-reports with a standardized, within-session proxy metric for comparative fragrance evaluation. Full article

Rapid sterility testing of radiopharmaceuticals is essential due to their short half-lives and strict safety requirements. Conventional culture-based methods require several days and are not compatible with clinical workflows. In this work, we present a proof-of-concept droplet-based microfluidic platform for rapid optical detection of bacterial contamination through optical extinction analysis of microdroplets. Monodisperse water-in-oil microdroplets were generated and optically interrogated using a fiber-based detection system. Calibration was first performed using 500 nm polystyrene nanoparticles to establish the relationship between particle concentration and optical extinction. Subsequently, Staphylococcus aureus suspensions were analyzed under aerobic and anaerobic conditions at concentrations ranging from 0 to 230 CFU/mL. The system demonstrated reliable detection of bacterial contamination with estimated limits of detection of ~15 CFU/mL (aerobic) and ~7.5 CFU/mL (anaerobic). The platform enables real-time, high-throughput analysis with minimal sample handling and reduced analysis time compared to conventional sterility tests. This study validates the feasibility of microdroplet-based optical detection as a rapid quality control strategy specifically suited for radiopharmaceutical production, where the short half-lives of common radiotracers impose strict time constraints incompatible with conventional 14-day culture-based sterility tests. The results provide a proof-of-concept foundation for future integration into automated sterility testing workflows, with further validation on real radiopharmaceutical matrices planned as the next step. Full article

Despite its significant water-saving potential, the adoption of alternate wetting and drying (AWD) irrigation remains limited due to infrastructure constraints and intensive manual monitoring requirements. An automated precision irrigation system was developed and tested at the Bangladesh Rice Research Institute research farm in Gazipur, Bangladesh. The system combined ultrasonic water-level sensors, capacitive soil moisture sensors, an Arduino-based microcontroller, a GSM communication module, and solar-powered automatic control gates. Field performance was evaluated following a Randomized Complete Block Design (RCBD) under four irrigation treatments: IRRISAT, IRRI35, IRRI25, and continuous flooding (CF). The first three irrigation treatments were operated using scheduled daily decision windows, in which irrigation actions were automatically triggered based on predefined schedules and sensor threshold values. In IRRISAT, irrigation started when soil moisture dropped slightly below saturation and stopped at a ponding depth of 5 cm, while IRRI35 and IRRI25 were triggered at volumetric soil water contents of 35% and 25%, respectively, with the same upper cutoff of 5 cm ponding depth; CF served as the control. The IRRI35 treatment achieved a high grain yield (7.76 t ha⁻¹) while reducing water use by 28% and energy consumption by 37% compared to CF. Water use efficiency was considerably higher under IRRI35 (9.4 kg ha⁻¹ mm⁻¹) than under CF (6.7 kg ha⁻¹ mm⁻¹). The automated system proved to be reliable and precise in scheduled irrigation control, significantly reducing water use and labor requirements. The findings suggest that large-scale adoption of the system under real-world cultivation conditions could reduce irrigation energy needs and contribute to sustainable water governance in rice production. Full article

Background: Perineural catheter migration is a clinically relevant cause of continuous block failure, but the present study was not designed to model true clinical displacement. Instead, we investigated whether low-rate infusion pressure differs between two predefined catheter–tissue environments, interfascial and intramuscular, under controlled ex vivo conditions. Methods: Sixty porcine thigh specimens were studied. Under ultrasound guidance, a catheter-over-needle system with a multi-orifice catheter was placed either in the interfascial plane or intramuscularly, with one measurement obtained from each specimen. After baseline recording outside the tissue, saline was infused at 5 mL/h for 10 min. Pressure recordings were normalized to baseline. For each trace, a representative value was obtained using a predefined automated stable-segment algorithm, and between-group differences were assessed using Welch’s t-test. Results: Mean normalized pressure was higher during intramuscular than interfascial infusion (0.3346 ± 0.0635 PSI [17.3 ± 3.3 mmHg] vs. 0.1917 ± 0.0285 PSI [9.9 ± 1.5 mmHg]). The between-group difference was significant (mean difference: 0.1430 PSI [7.4 mmHg], 95% CI: 0.1181 to 0.1679 PSI; p = 7.22 × 10⁻¹⁵), with a very large standardized effect size (Hedges’ g = 2.87), reflecting strong statistical separation between the two predefined groups under controlled ex vivo conditions rather than clinical discriminative ability. However, the absolute pressure difference remained small. Conclusions: Under controlled ex vivo conditions, mean normalized infusion pressure differed between predefined interfascial and intramuscular catheter positions. However, the absolute difference was small. This binary model does not represent real catheter displacement, and the findings do not support current clinical applicability, individual-level interpretation, or the definition of a clinically usable threshold. The results should be considered exploratory and hypothesis-generating. Full article

This article assesses the potential for intra-day redistribution of the electrical load of water intake systems under different electricity tariff models, using water supply systems in Belarus and China as case studies. It demonstrates how tariff policy influences the electrical load profile of a water intake system and quantitatively evaluates the economic effect of optimizing the operating modes of pumping equipment. The analysis is based on daily profiles of electric power and water supply. For the Belarusian water supply system, data for 2019 were considered, corresponding to the baseline operating mode without targeted load management, and data for 2023 were considered after the transition to dispatch-based control of well activation with account taken of tariff constraints (without automation tools). For the Chinese water intake system, hourly data for 2025 were used. The load redistribution potential was assessed on the basis of lagged correlation between power and water supply profiles. In addition, the F-index was applied as an aggregated diagnostic indicator intended for the comparative assessment of potential load transferability across technological stages, taking into account their share in total energy consumption. For the Chinese case, it was shown that the maximum correlation between water supply and electricity consumption across all technological stages is achieved near zero lag, which indicates a high adaptation of system operating modes to current demand; at the same time, the R values were 0.19 for reservoir intake, 0.86 for water treatment, and 0.51 for the pumping station. In the Belarusian case, for the first-lift stage, the maximum correlation is shifted by −6 h relative to zero lag, indicating a less rigid linkage of pump operation to current demand and a more inertial response of the system. A comparison of 2019 and 2023 for the Belarusian facility showed that targeted regulation of well activation and load redistribution across tariff zones reduced the total electricity cost by 1.58%, confirming the potential for further optimization of electricity consumption regimes. Full article

Unlike reactive systems, mechanical ventilation controlled by CO₂ concentration operates at a target efficiency that dynamically increases whenever the target CO₂ level is exceeded. This approach eliminates the typical ‘dead-time’ and prevents air quality degradation by ensuring the system adjusts its performance immediately in response to concentration changes. In this work, the study focuses on the development and evaluation of data-driven predictive models for near-term indoor CO₂ forecasting that can be integrated into pre-occupancy ventilation strategies, rather than designing a complete control scheme. Experimental data were collected over four months in a 48 m² smart laboratory configured as an open-plan office, where a heterogeneous IoT sensing architecture logged synchronized time-series measurements of CO₂ and microclimate variables (temperature, relative humidity, PM_2.5, TVOCs), together with acoustic noise levels and appliance-level energy consumption used as indirect occupancy-related signals. Raw telemetry was transformed into a 22-feature state vector using a structured feature engineering method incorporating z-score standardization, cyclic time encodings, multi-horizon CO₂ lags, rolling statistics, momentum features, and non-linear interactions to represent temporal autocorrelation and daily periodicity. The study benchmarks multiple regression paradigms, including simple baselines and ensemble methods, and found that an automated multi-level stacked ensemble achieved the highest predictive fidelity for short-term forecasting, with an Mean Absolute Error (MAE) of 32.97 ppm across an observed CO₂ range of 403–2305 ppm, representing improvements of approximately 24% and 43% over Linear Regression and K-Nearest Neighbors (KNN), respectively. Temporal diagnostics showed strong phase alignment with observed CO₂ rises during occupancy transitions and statistically reliable prediction intervals. Five-fold walk-forward cross-validation confirmed the temporal stability of these results, with top models achieving consistent R² values of 0.93–0.95 across Folds 2–5. These results demonstrate that, within a single-room university laboratory setting, historical sensor data from low-cost IoT devices can support accurate short-term CO₂ forecasting, providing a predictive layer that could support future proactive ventilation scheduling aimed at reducing CO₂ lag at the start of occupancy while avoiding unnecessary ventilation runtime. Generalization to other building types and occupancy profiles requires further validation. Full article

Home automation is undergoing a paradigm shift from connected IoT environments with rule based control to intelligent homes exhibiting ambient intelligence and proactive adaptation. Artificial intelligence, privacy-preserving sensing, and converging connectivity standards are the primary forces driving this transition. This systematic literature review synthesizes the technological foundations, architectural developments, emerging paradigms, and socio-technical challenges characterizing the next generation of smart homes, evaluated against the original Ambient Intelligence (AmI) vision. Following PRISMA 2020 guidelines, searches were conducted across four databases—IEEE Xplore, ACM Digital Library, Scopus, and Web of Science—covering studies published between January 2020 and June 2025. From 3450 records, 113 studies were selected through a two-reviewer screening procedure with inter-rater reliability assessments. Quality was assessed using a modified JBI Critical Appraisal Checklist, and findings were synthesized through thematic analysis. Three converging technological pillars were identified: multi-modal privacy-preserving sensing including mmWave radar; a hierarchical cloud-edge TinyML intelligence engine; and unified connectivity through the Matter/Thread standard. Emerging paradigms include LLM-based cognitive orchestration, hyper-personalization, Digital Twin simulation, and grid-interactive prosumer energy management. Realizing that the intelligent home vision requires addressing the privacy–security–trust trilemma, algorithmic bias, system reliability, and human–agent collaboration, a research roadmap encompassing explainable AI, privacy-by-design, lifelong learning, and standardized ethical auditing is proposed. Full article

The transition from traditional synchronous generators to intermittent renewable sources, combined with increasingly variable and difficult-to-control energy demand, is creating a growing need for large-scale reserves and energy storage. At the same time, reduced system inertia and evolving electricity market regimes are emerging as important challenges that may affect grid stability, reliability, and economic performance. Advanced storage technologies, particularly those with fast ramping and high-response capabilities, offer a potential means of providing near-instantaneous support in response to unexpected system disturbances or market signals, thereby helping to mitigate inertia-related risks. This paper investigates four technologies: pumped hydroelectric storage, battery energy storage systems, synchronous condensers, and special protection schemes, with a focus on their capability to deliver rapid responses to large-scale disturbances. The analysis is conducted using a deliberately simplified power system model to provide qualitative insights into system behavior and control interactions. The results indicate that automation-enabled responses to system imbalances, including support from synchronous condensers and the rapid activation of additional generation, can enhance system performance under disturbance conditions within the considered framework. These findings demonstrate the feasibility and potential value of such approaches; however, further validation using higher-fidelity models and system-specific data is required to quantify their operational and economic impacts. Full article

The multi-finger robotic hand offers great potential for precise control due to its high degrees of freedom. Yet, manipulating objects forms a closed-chain kinematic system, which compounds the dimensionality and computational complexity of trajectory tracking. To tackle this challenge, and inspired by the widespread application of the zeroing neurodynamics (ZND) in robotic control, this study proposes a novel direct-discrete robust neurodynamics (DDRN) algorithm. The proposed algorithm advances the ZND methodology by employing a direct discretization design strategy. This strategy is crucial for two reasons. First, it fits naturally with the discrete-time nature of digital systems, enabling practical implementation. Second, it enhances precision by avoiding the integration errors inherent in continuous-to-discrete transformations. By simultaneously integrating this direct discretization with explicit noise suppression mechanisms, the DDRN algorithm efficiently solves the high-dimensional tracking problem formulated as a constrained time-varying quadratic programming (CTVQP) problem. Theoretical analyses demonstrate that under various noise environments, the steady-state residuals (SSRs) achieve global convergence, guaranteeing the algorithm’s strong robustness and high accuracy. Furthermore, comprehensive numerical simulations substantiate its superior performance. Practically, this DDRN algorithm enables more reliable and precise real-time control of dexterous robotic hands, with potential benefits for advanced manufacturing, prosthetic hands, and automated assembly where accurate trajectory tracking under sensor noise is critical. Full article