Search Results (13,178)

Many hands-on tasks remain difficult to fully automate because they require human dexterity and flexible object handling. Data gloves offer a promising interface for sensing hand–object interactions, but most prior systems focus on gesture recognition or object classification rather than closed-loop, step-by-step task guidance. In this work, we develop and evaluate a tactile-sensing operation support system using an e-textile data glove with 88 pressure sensors, a tactile pressure sheet for placement verification, and a GUI that provides step-by-step instructions. As a core component, a CNN classifies the grasped state as bare hand or one of four discs with 93.3% accuracy using 16,175 training samples collected from five participants. In a user study on the Tower of Hanoi task as a controlled proxy for multi-step manipulation, the system reduced mean solving time by 51.5% (from 242.6 s to 117.8 s), reduced the number of disc movements (35.4 to 15, about 20 fewer moves on average), and lowered perceived workload (NASA-TLX) by 53.1% (from 68.5 to 32.1), while achieving a SUS score of 75. These results demonstrate the feasibility of tactile-based step verification and guidance in a controlled multi-step task; broader generalization requires evaluation with larger and more diverse participant groups and tasks. Full article

The integration of automated learning and video analysis enables the development of intelligent systems that can operate effectively in uncertain scenarios. These systems can autonomously identify dominant motion dynamics, depending on the theoretical framework used for representation and the learning process used for pattern identification. Current literature offers a state-based approach to describe the key temporal and spatial relationships required to understand motion dynamics. An important aspect of this approach is determining when the number of positively learned rules from a given information source is sufficient to detect dominant motion in automatic surveillance scenarios. This is crucial, as it affects both the variability of movements that monitored subjects can exhibit within the camera’s field of view and the resources needed for effective implementation. This study addresses these gaps through a grammar-based sufficiency criterion, which posits that learning is complete when production rule growth stabilizes, under the assumption of system stationarity. The stability criterion evaluates whether the most probable rules are learned over time, and whenever a high-growth rule is added, it is used to update the criterion. We outline several benefits of having a formal criterion for determining when a symbolic surveillance system has a robust model that explains the observed motion dynamics. Our hypothesis is that a correct model can consistently account for the majority of motion dynamics over time in an automated learning process. The proposed approach is evaluated by modeling motion dynamics in several scenarios using the SEQUITUR algorithm as input and computing the probability of stability along the learning curve, which indicates when the model reaches a steady state of consistent learning. Experimental validation was conducted in real-world scenarios under varying acquisition conditions. The results show that the proposed method achieves robust modeling performance, with accuracy values ranging from 83.56% to 95.92% in dynamic environments. Full article

Data-driven approaches to blood glucose predictive modeling face significant challenges due to the inherent variability in biological systems. While these methods efficiently capture statistical patterns through automated processes, they often lack the biological interpretability necessary to link model behavior with underlying physiological mechanisms. In contrast, physiological models offer accurate mechanistic representations but require complex parameterization and specialized domain expertise. In this work, we present an approach for predicting blood glucose levels (BGLs) leveraging the concept of physiology-informed neural networks (PINNs). This approach addresses the challenge of BGL prediction by incorporating the parameters of insulin and meal dynamics within the architecture of a predictive network. It employs a two-stage learning approach for modeling physiology and predicting BGLs. The neural network is pretrained to approximate the solutions of the physiological dynamics, and the output of this pretrained model, representing the insulin and glucose concentration states, is then fed as input into a predictive model, enabling simultaneous optimization of predictive accuracy and physiological parameter estimation, offering advantages over traditional modeling approaches in terms of personalized prediction and interpretability. The results highlight the model’s ability to estimate physiological parameters while maintaining strong predictive performance that aligns with the underlying physiological principles. This framework offers significant potential for personalized predictive modeling where precise and efficient understanding of individual metabolism is essential. Full article

Manual harvesting of white button mushrooms involves coordinated bending and twisting motions to detach the fruiting body while minimizing surface damage; however, replicating these actions in automated systems remains challenging. In this study, a vacuum-based end-effector that mimics manual twist–bend detachment using a single-point contact was designed and evaluated to reduce mechanical damage. Key detachment parameters, including the friction coefficient (mean 0.62), bending angle (average 5.72°), and twisting torque (average 2.56 N·m), were experimentally analyzed to determine the minimum vacuum pressures required for effective bending and twisting, which were −8.64 ± 2.21 kPa and −8.91 ± 2.45 kPa, respectively, with no significant difference observed between the two motions (p = 0.51). A customized vision-based image processing algorithm was developed to quantify postharvest surface damage using a whiteness index (WI). An optimal vacuum pressure of −17.17 kPa was identified, together with a bending angle of 10° and a twisting angle of 90°, balancing high harvesting success with preservation of mushroom quality. The results highlight the influence of end-effector design parameters, including vacuum cup material, contact area, bending direction, and vacuum application duration, on harvesting performance and product marketability, supporting the development of robotic systems for fresh mushroom harvesting. Full article

We developed a real-time alert and data management framework that integrates programmable logic controllers, RS-485 industrial communication, Structured Query Language Server, Message Queuing Telemetry Transport (MQTT), and the nodemation (n8n) automation platform, using a filling machine production line as a case study. The system collects and analyzes the operational status and production line data of the filling machine in real time, storing all information in a database for preservation. Through MQTT, the data is sent to n8n for automated processing. When equipment anomalies occur or data exceed predefined thresholds, the system automatically notifies maintenance personnel via communication software APIs. Additionally, users can query daily production capacity or related data using n8n’s AI functions. This architecture offers low cost, rapid deployment, cross-platform integration, and high flexibility. It not only improves anomaly handling efficiency but also preserves complete historical records, supporting trend analysis, report generation, and decision optimization, thereby assisting the filling production line in achieving long-term stable and intelligent management. Full article

Automated pig-welfare monitoring needs scalable, non-invasive signals that work across ages and individuals. A key methodological contribution of this study is the use of subject-wise validation, which ensures generalization to unseen animals and prevents inflated accuracy caused by growth-related and individual ‘voice’ differences. Vocalizations can help, but growth and individual “voice” differences can confound distress patterns and overstate accuracy without subject-wise validation. In our study, we explicitly accounted for individual variability by including animal identity as a random effect in mixed models and by using grouped cross-validation, where models were tested only on pigs not seen during training. This approach ensures that the reported accuracy reflects generalization across different individuals rather than memorization of specific vocal signatures. We analyzed 2221 vocal samples from 40 pigs (20 males, 20 females) recorded across four growth phases (farrowing, nursery, growing, finishing) under six conditions (pain, hunger, thirst, cold stress, heat stress, normal). Acoustic features extracted in Praat included energy, duration, intensity, pitch, and formants (F1–F4). Using blockwise variance decomposition, we quantified contributions of distress exposure, growth phase, and sex, and estimated the additional variance explained by animal identity. Distress exposure dominated intensity and spectral traits, particularly Formant 2, whereas the growth phase produced systematic shifts in duration and pitch. Animal identity added a modest but consistent increment in explained variance (~+0.02–0.03 R² beyond sex, phase, and distress). For prediction, we used 5-fold cross-validation grouped by animal. A Random Forest achieved a modest balanced accuracy of 0.609 and macro-F1 of 0.597; pain was most separable (recall 0.825), while other states showed moderate recall, indicating overlap. These results support hierarchical acoustic encoding of distress and establish a benchmark for precision welfare monitoring. Furthermore, they highlight that resolving complex physiological overlaps, such as heat stress and resource competition, requires a shift from unimodal acoustic models to multimodal Precision Livestock Farming (PLF) systems that integrate bioacoustics with continuous environmental and behavioral data streams. Full article

Post-harvest losses of fruits and vegetables are an important bottleneck in food systems of countries around the world, with 30–50% of perishable food items lost between farm and consumer, smallholder farmers in low-and-middle income countries (LMICs) with poor cold chain infrastructures facing a disproportionate burden. Evaporative cooling (EC) is a low-cost and energy-efficient alternative to mechanical refrigeration; however, traditional systems are operated in one position and are dependent on climate, which restricts its performance. The combination of Internet of Things (IoT) sensing, machine learning (ML), and the advanced control theory has made intelligent evaporative cooling systems (IECS) adaptive, data-driven platforms that can regulate the environment in real-time and optimise autonomously. This is a systematic literature review that was carried out according to PRISMA 2020, summarising 94 peer-reviewed articles published in 2018–2025 to map the technological landscape, performance indicators, and research directions of the field of post-harvest fruit and vegetable preservation using IECS. Findings indicate that IECS can considerably lower the storage temperatures, increase the shelf life by 50–200%, and reduce energy consumption by 75–90% compared to traditional refrigeration, and the payback period is as short as 1.2 years. In arid conditions, ML models are accurate in prediction with an R² of 0.98. The gaps in the research identified are a lack of validation in wet climatic conditions, non-existent standardised Ag-IoT protocols, inadequate Food–Energy–Water (FEW) nexus calculation, and no explainable AI (XAI) interfaces. An example of a conceptual framework of four layers synthesised is proposed to direct next-generation research and implementation of the IECS. Full article

A water company’s water supply network is responsible for distributing good-quality water in quantities that meet customer needs, ensuring proper operation of the water supply network to ensure adequate pressure at the receiving points, efficiently repairing faults, and planning and executing maintenance, modernization, and expansion work. Managing a water supply network is a complex and complex process. A crucial challenge in water company management is detecting and locating hidden water leaks in the water supply network. Leak location in water distribution networks is a key challenge for utilities, as undetected leaks lead to water losses, increased energy consumption, and reduced service reliability. With the development of cyber-physical systems (CPSs), the integration of physical infrastructure with real-time digital monitoring has enabled more adaptive and responsive water operations. Data-driven decision-making in CPS in the water industry leverages classification and regression trees (C&RTs) to analyze real-time sensor data—such as pressure, flow, and consumption—to classify system states and predict potential faults. By transforming operational data into interpretable decision rules, C&RTs enable automated and timely maintenance actions that improve reliability, reduce water loss, and support intelligent infrastructure management. The aim of this study is to develop and evaluate AI-based optimization methods to enhance sustainability, efficiency, and resilience in the water industry by enabling autonomous, data-driven decision-making within CPSs, supporting smart industry practices, and addressing practical challenges associated with the actual implementation of smart water management solutions using simple solutions such as C&RTs. The accuracy of the best classifier was 86.15%. Further research will focus on using other types of decision trees that will improve classification accuracy. Full article

The development of novel antiepileptic agents requires early identification of pharmacokinetic limitations to mitigate risks at later stages. This study aimed to perform in silico profiling of a library containing 448 novel 2H,5H-chromeno[4’,3’:4,5]thiopyrano[2,3-d]thiazol-2-one derivatives to select lead compounds with an optimal balance of safety and efficacy. The study was conducted using the ADMET-AI platform, based on a graph neural network, to predict physicochemical, pharmacokinetic, and toxicological properties. The methodology involved calculating drug-likeness descriptors for primary screening and a comparative statistical analysis of the top 20 selected structures against 16 approved antiepileptic drugs and four reference compounds. Based on drug-likeness descriptors and predicted ADMET (absorption, distribution, metabolism, excretion, toxicity) related parameters, 20 structures were prioritized for further analysis. Their predicted profiles suggested high intestinal absorption and blood–brain barrier (BBB) permeability, which may be relevant for central nervous system (CNS) directed agents. In comparison with the reference thiazolidinones, the prioritized compounds showed comparatively more favorable predicted mutagenicity and carcinogenicity profiles. Elevated predicted risks of hepatotoxicity and cardiotoxicity were observed for several structures, indicating the need for further structural optimization. The results suggest that the thiopyranothiazolidinone scaffold merits further anticonvulsant-oriented investigation at the stage of early compound prioritization. Experimental validation will be required to confirm the actual pharmacokinetic, toxicological, and anticonvulsant properties of the prioritized compounds. Full article

This is the first systematic review evaluating Black Sea plankton as biosensor organisms for Biological Early Warning Systems (BEWS)—real-time monitoring approaches that detect sublethal behavioral or physiological responses to pollutants before irreversible ecosystem damage occurs. The systematic literature review was guided by the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) approach, ensuring methodological transparency and applicability. A total of 140 publications from databases (Web of Science Core Collection, Scopus, PubMed, and Google Scholar databases) were included in the final analysis. We assess nine native planktonic taxa as candidates for automated video-based water quality monitoring, using a multi-criteria framework encompassing biological sensitivity, technical detectability, and practical feasibility. Three species emerge as the most suitable candidates: Aurelia aurita as a universal indicator (sensitive to copper, surfactants, petroleum, and microplastics; its large size enables standard video detection); Acartia tonsa for trace contamination (reproductive toxicity at metal concentrations 4–33× below regulatory standards); and Mnemiopsis leidyi for metal-specific discrimination (bioluminescent responses: 650% Zn, 430% Cu, and 350% Hg at 0.001 mg/L). Analysis of 140 publications reveals critical gaps: 33% of species lack toxicological data, 95% of studies test single toxicants despite natural mixture exposure, and microplastic methodology varies 1000-fold in particle size. Threshold analysis suggests planktonic sublethal stress at “safe” concentrations under current standards, suggesting inadequate protection of marine food webs. A complementary monitoring approach integrating these species with computer vision algorithms offers autonomous early-warning capability for Black Sea environmental management. Full article

Accurate and scalable body condition scoring (BCS) is important for health monitoring and productivity management in precision livestock farming. However, manual scoring is subjective, labor-intensive, and difficult to standardize, while many automated methods are too computationally demanding for edge deployment in real farm environments. This study proposes EdgeBCS-YOLO, a lightweight object detection framework for real-time beef cattle BCS in unstructured farming scenarios. Built on YOLO11n, it combines Position-Sensitive Feature Fusion (PSFF), a Texture-Aware Star Module (TASM), an Efficient Grouped Detection Head (EGDH), and a Focal and Global Knowledge Distillation (FGD)-based distillation strategy. On a dynamic blurring dataset, EdgeBCS-YOLO achieved 90.8% precision, 82.7% recall, and 88.9% mAP@50. On the NVIDIA Jetson Orin NX Super, it achieved a model size of 3.95 MB, a system FPS of 33.35, and an average inference latency of 13.26 ms. These results suggest that it is a practical and potentially efficient solution for automated BCS on edge devices. Full article

Patients leave millions of medical reviews annually, providing critical data for quality management. However, manual processing is infeasible, and existing systems fail to distinguish medical from organizational problems—a distinction essential for complaint routing. The consequences of misrouting are significant: clinical issues may go unaddressed when medical complaints reach administrative staff, while systemic service problems remain unresolved when organizational complaints reach medical directors. We developed a hybrid approach combining expert annotation with Large Language Models (LLMs). Fifteen prompt iterations on 1500 reviews with expert validation (modified Cohen’s kappa (κ_mod), which weights errors hierarchically, reached 0.745) preceded the LLM annotation of 15,000 mixed-sentiment and positive reviews. These were combined with 7417 expert-annotated negative reviews to form a corpus of 22,417 reviews. Eight architectures, ranging from Logistic Regression to a BERT + TF-IDF + LightGBM ensemble, were compared using both standard metrics and domain-specific practical metrics tailored to complaint routing. The best model, scaled to 4.3 million Russian-language reviews from the Prodoctorov.ru platform, achieved 92.9% Practical Accuracy—the proportion of reviews classified without critical medical–organizational misclassification errors (M ↔ O)—compared to 68.0% standard accuracy, which treats all errors equally. Critical errors were reduced to 1.4%, yielding 144,000 more correctly processed complaints than traditional methods (TF-IDF + Logistic Regression). Analysis of the scaled data revealed the following: 46.1% M (medical), 21.0% O (organizational), and 32.9% C (combined) reviews; medical ratings were highest (4.75 vs. 4.59 for organizational, p < 0.001); combined reviews were longest (802 characters); zero-star reviews comprised 3.8% of feedback, with organizational complaints dominating (38.2%) among extreme negatives; and average ratings rose by 1.24 points over 14 years. This hybrid approach yields expert-comparable corpora, automates 93% of feedback processing, ensures correct complaint routing, and contributes to healthcare sustainability by reducing administrative burden, accelerating resolution, and enabling data-driven quality management without proportional increases in human resources. All analyses were conducted on Russian-language patient reviews. Full article

Beekeeping is a profitable and mind-relaxing practice; however, monitoring beehives poses significant challenges, such as consuming time and potentially disturbing hive equilibrium, which may lead to colony collapse. Developing precision beekeeping (PB) systems is crucial to assist beekeepers in decision-making, automate redundant hive maintenance, and enhance the security and comfort of bee life. This review systematically explores research on PB systems, based on a keyword-driven search of Scopus and Web of Science databases, yielding 46 relevant publications. The analysis highlights a notable increase in research activity in the field since 2016. The integration of advanced technologies, including machine learning, cloud computing, IoT, and scenario-based communication methods, has proven instrumental in predicting hive states such as queen status, enemy attacks, readiness for harvest, swarming events, and population decline. Commonly measured parameters include hive weight, temperature, and relative humidity, with various sensors employed to ensure precision while minimizing bee disturbance. Additionally, bee traffic monitoring has emerged as a critical approach to assessing hive health. Most studies focus on honeybees rather than stingless bees and, in the context of enemy identification, Varroa destructor is the primary target. This review underscores the potential of novel technologies to revolutionize apiculture and enhance hive management practices. Full article

Urban socio-economic resilience to climate change has emerged as a central research theme as cities increasingly confront interconnected environmental, economic, and social risks. Despite the rapidly expanding body of literature, the conceptual boundaries, thematic evolution, and analytical priorities of this field remain fragmented across disciplines, and no prior study has systematically mapped the socio-economic dimension of urban resilience through a combined bibliometric and thematic analysis over a multi-decadal horizon. This study addresses that gap by providing a systematic review of global research on urban socio-economic resilience to climate change, integrating bibliometric and thematic analyses of peer-reviewed publications from 1990 to 2025. Following the PRISMA 2020 guidelines, records were retrieved from the Web of Science Core Collection and subjected to a multi-stage screening procedure that combined automated relevance scoring with mandatory manual validation of the socio-economic dimension, resulting in a final dataset of 5076 publications. The analysis examines conceptual interpretations of socio-economic resilience, dominant climate hazards affecting urban systems, methodological approaches and assessment indicators, adaptation strategies and governance responses, and emerging research gaps. The results reveal a marked acceleration of scientific output after 2015, driven by the Paris Agreement and the IPCC Special Report on Global Warming of 1.5 °C (2018). The bibliometric network analyses identify adaptation, vulnerability, flooding, and sustainability transitions as the core thematic clusters. The findings trace a paradigmatic trajectory from equilibrist recovery frameworks toward transformative, socio-economically grounded resilience models and reveal persistent gaps in the operationalization of governance, equity measurement, and geographic representation. By synthesizing three-and-a-half decades of scholarship, this review clarifies the intellectual structure of the field and proposes four specific post-2026 research pathways that emphasize longitudinal cross-city comparisons, mixed-methods assessments, sector-specific compound hazard analyses, and governance mechanism studies. Full article

Background/Objectives: Totally endoscopic mitral valve repair reduces surgical trauma and accelerates recovery but can be technically challenging, particularly for precise annuloplasty suturing. The VirtuoSEW^® (LSI Solutions, Victor, NY 14564m, USA) automated annular suturing system was developed to standardize and simplify suture placement. This study was an early evaluation of this technology’s safety, efficacy, and feasibility in totally endoscopic microInvasive mitral valve repair (µMVr). Methods: We conducted a retrospective observational study of 20 patients with severe mitral valve disease of various etiologies. All patients underwent mitral valve repair using the VirtuoSEW^® system for automated placement of annuloplasty sutures, combined with leaflet resection or chordal management as appropriate. Postoperative outcomes were assessed at one month using echocardiography and clinical evaluation. Perioperative and postoperative complications and early mortality were systematically recorded. Results: VirtuoSEW^®-assisted mitral valve repair was safe and effective, achieving complete elimination of severe mitral regurgitation in all patients (N = 20, 100%). Annuloplasty rings included Physio-ring (N = 12, 60%), Memo 3D (N = 4, 20%), and Memo 4D (N = 4, 20%), combined with leaflet repair techniques: leaflet plication (N = 5, 25%), neochordae implantation (N = 7, 35%), sliding plasty (N = 2, 10%), commissural repair (N = 1, 5%), and hemibutterfly repair (N = 1, 5%). Concomitant procedures included: tricuspid valve repair (N = 1, 5%) and atrial septal defect closure (N = 1, 5%). Mitral annulus diameter decreased from 42.0 ± 5.3 mm to 34.2 ± 2.2 mm (p = 0.001). Mean total surgery, cardiopulmonary bypass, and aortic cross-clamp times were 170.3 ± 21.3, 143.4 ± 21.5, and 80.4 ± 7.9 min, respectively. ICU stay was 1.0 ± 0.2 days, with a hospital stay of 8.0 ± 1.9 days. No perioperative complications—including bleeding (N = 0, 0%), stroke (N = 0, 0%), infections (N = 0, 0%), or 30-day mortality (N = 0, 0%)—occurred. Conclusions: µMVR invasive mitral valve repair using the VirtuoSEW^® system is safe, effective, and reproducible, as well as compatible with almost all repair techniques, providing complete restoration of valve competence with no early device-related complications. To our knowledge, this is the first clinical study reporting outcomes with this device, supporting its potential to streamline mitral repair and improve procedural efficiency. Full article