Search Results (70)

The post-pandemic evolution of education involving mechatronics and machine learning has shifted the demand for robotic hardware from centralized laboratories to accessible laboratories in home environments. This paper presents a portable three-wheeled holonomic robotic platform designed for remote research and home office experimentation. The proposed system utilizes a modular design and low-cost philosophy comprising a custom embedded control system driven by an ESP32-WROOM microcontroller, which manages a closed-loop PID velocity controller using Hall effect feedback from three DC micromotors. In contrast, external nodes allow the reception, conditioning, and classification of 8-channel surface electromyography (sEMG) data sampled at 500 Hz. To address the non-stationarity and stochastic noise in raw sEMG signals, this study implements a hybrid Deep Learning (DL) architecture that complements 2D Convolutional Neural Networks (CNN) for spatial feature extraction with Long Short-Term Memory (LSTM) networks for temporal context awareness. This model decodes the neuromuscular intent of the user into real-time holonomic velocity vectors, achieving validation accuracies of 80.51% for horizontal movement, 84.86% for vertical translation, and 99.56% for the Fist/no-Fist state. By synthesizing advanced AI-based teleoperation with a portable design, this study establishes a scalable framework for the next generation of “laboratory-at-home” educational tools and research regardless of physical location. Full article

Baijiu, a traditional Chinese distilled spirit with profound cultural and economic significance, faces long-standing challenges in standardization, quality consistency, and skill inheritance due to its empirical production model. The rapid advancement of artificial intelligence (AI) and multi-omics technologies is driving a paradigm shift in Baijiu research from experience-driven to data-driven approaches. This review systematically summarizes the current state of AI applications across the entire Baijiu industry chain. Common AI methods including traditional machine learning, deep learning, multimodal data fusion, and emerging paradigms such as explainable AI (XAI), genome-scale metabolic models (GEMs), and few-shot learning are critically examined. Key bottlenecks—data silos, small sample sizes, model interpretability, and the tension between technology and tradition—are discussed in depth. Future directions are proposed, including multimodal fusion, digital twins, hybrid mechanistic–data modeling, closed-loop control, human–machine collaboration, standardization, and ethical governance. This review provides a comprehensive framework for integrating AI into Baijiu research and offers references for intelligent transformation in other fermented food systems. Full article

Artificial intelligence (AI) is rapidly redefining the management of urban energy systems by coupling predictive analytics with closed-loop control across buildings, power grids, and mobility networks, positioning cities as critical leverage points in global decarbonization efforts. Contemporary urban environments generate vast, heterogeneous datasets that enable advanced machine learning applications; however, limitations remain, including interpretability–fairness trade-offs, fragmented data governance, interoperability gaps, cybersecurity risks, and insufficient long-term validation across diverse climatic and socio-economic contexts. This review evaluates AI-driven strategies for energy-efficient urban systems and identifies the technical and governance conditions required for scalable impact. The evidence synthesized indicates that supervised and ensemble learning models achieve high predictive accuracy for electricity demand and chiller performance, with models such as Random Forest Regression achieving R² values up to 0.9835 in electricity consumption forecasting, while unsupervised approaches detect latent inefficiencies in HVAC systems, delivering measurable savings typically around 6% under controlled benchmarking conditions. Deep learning architectures improve multi-building forecasting and real-time control, with hybrid CNN–LSTM models achieving prediction accuracies up to 97% and outperforming traditional statistical approaches in weekly energy demand forecasting achieving higher prediction accuracy and significant energy savings in complex urban subsystems with reported reductions of approximately 21–23% in residential and educational buildings and up to 37% in office HVAC systems. Hybrid and physics-informed AI models embed thermodynamic principles into data-driven frameworks, improving robustness, interpretability, and generalization. IoT sensor networks and edge-computing architectures support adaptive HVAC, demand–response, and smart-grid management, while integrated building–grid–mobility systems enhance load balancing, storage use, and carbon reduction. AI-enhanced strategies offer a credible pathway toward measurable reductions in urban energy use and emissions with deep reinforcement learning in digital twin environments reducing HVAC energy demand by 10–35% while maintaining thermal comfort within ±0.5 °C in line with ASHRAE standards, and overall energy savings reaching up to 44% in optimized systems when supported by interoperable infrastructures, secure digital-twin architectures, and standardized measurement and verification protocols. Full article

This study proposes EvoDeep-Quality, a closed-loop hybrid framework integrating deep learning-based perception with multi-objective evolutionary optimization for adaptive quality control in smart manufacturing. The architecture combines a CNN-LSTM network for real-time spatiotemporal quality prediction with an NSGA-III-based optimization unit to balance conflicting objectives of quality, cost, and energy efficiency. A continuous adaptive learning loop addresses concept drift and process variability. Evaluated on an industrial-inspired synthetic dataset of textile blends (N = 5000) and validated on the real-world SECOM semiconductor manufacturing dataset, the framework demonstrates strong predictive capability (R² = 0.947 ± 0.012, MAE = 0.035 ± 0.003) and significant manufacturing performance improvements, including a 23.5% quality enhancement and an 8.7–12.3% operational cost reduction compared to traditional and standalone AI models. Statistical significance testing (paired t-test, p < 0.01) confirms the superiority of the proposed approach. This deep-evolutionary framework advances proactive quality assurance and adaptive process control, offering a scalable solution aligned with Industry 4.0 and 5.0 paradigms. Full article

Metal packaging materials remain fundamental across food, beverage, pharmaceutical, cosmetic, and technical sectors owing to their combination of mechanical robustness, total light and gas barrier performance, thermal resistance, and established recyclability. Aluminum alloys, tinplate, tin-free steel (TFS/ECCS), stainless steels, metal–matrix composites (MMCs), and metal–polymer or metal–paper laminates define distinct metal-based packaging architectures whose metallurgical and interfacial design governs forming behaviour, corrosion and migration pathways, coating integrity, and mechanical reliability. In this review, these architectures are examined from a materials- and systems-oriented perspective, linking composition, microstructure, processing routes, and surface engineering to functional performance across rigid, semi-rigid, and flexible formats. The analysis also considers the ongoing transition from bisphenol A (BPA)-based epoxy linings to BPA-free and hybrid coating chemistries, the use of nano-structured metallic and metal-oxide surfaces, and the role of composite laminates in which thin metallic foils are combined with polymeric or paper-based structural layers. These material and architectural aspects are discussed together with safety, regulatory, and circularity considerations that increasingly influence the design and selection of metal-based packaging. Ion migration, coating degradation, and corrosion under realistic storage environments are considered in relation to EU, FDA, ISO, and sector-specific requirements, while attention is also paid to the contrast between well-established closed-loop recycling infrastructures for aluminum and steel and the more complex end-of-life management of coated metals and multilayer laminates. The review provides a unified framework connecting materials selection, metallurgical design, processing, performance, regulatory compliance, and sustainability in metal-based packaging systems. Applications spanning consumer goods, pharmaceuticals, cosmetics, and advanced electronics are integrated to support an overall understanding of how metallic and hybrid metal-based architectures underpin functional reliability and life-cycle sustainability. Full article

Objective: This paper seeks to provide an effective and automated method for the creation and updating of building information modeling compliance rules using the integration of human-in-the-loop collaboration with advanced natural language processing. Methods: We propose a hybrid approach that integrates BERT-based semantic extraction, CFG structural validation, and confidence-based expert review. Results: Tested on a real-world power infrastructure project, the framework was found to be 95.8% accurate in translation and 98.3% feasible in rule execution, outperforming benchmark automated approaches. The human effort was reduced by 90% (168 h vs. 1620 h), and processing of regulatory changes was sped up by 94%. Conclusion: Data analysis shows that collaborative intelligence is a significant factor in closing the semantic and pragmatic gap for regulatory compliance. Compared with fully automated “black box” approaches, this method supplies a tractable, manageable, and operationally valid solution, giving a competitive edge over existing digital construction methods. Full article

This paper presents an efficient optimization framework for high-frequency control of active downhole Flow Control Valves (FCVs) under geological uncertainty. Traditional proactive optimization methods for FCVs, while capable of maximizing life-of-field objectives such as Net Present Value (NPV), are computationally prohibitive when frequent updates are required. Conversely, reactive approaches are efficient but often neglect long-term recovery objectives. To address these challenges, we integrate two complementary strategies within a reservoir simulator: a reactive nonlinear programming method to maximize instantaneous cash flow, and a proactive streamline-based Time-of-Flight (TOF) equalization approach to improve sweep efficiency by balancing flood front arrival times. The framework is demonstrated on synthetic and realistic reservoir models, including the Olympus and Almakman references. Results show that, compared to conventional annual control strategies, the proposed approach increases NPV by 15–25% while reducing water handling costs and deferring breakthrough by up to four years. Furthermore, hybrid optimization effectively neutralizes fracture uncertainty, improving both mean recovery and the certainty of outcomes. Three field-scale case studies highlight the practical benefits of FCVs in improving lift performance, maximizing recovery from bypassed hydrocarbons, and reducing the number of wells required to meet production targets. By combining reactive and proactive control within a computationally tractable workflow, this study advances the practical deployment of intelligent completions for closed-loop reservoir management. Full article

Artificial intelligence (AI) is rapidly reshaping adult cardiac surgery, enabling more accurate diagnostics, personalized risk assessment, advanced surgical planning, and proactive postoperative care. Preoperatively, deep-learning interpretation of ECGs, automated CT/MRI segmentation, and video-based echocardiography improve early disease detection and refine risk stratification beyond conventional tools such as EuroSCORE II and the STS calculator. AI-driven 3D reconstruction, virtual simulation, and augmented-reality platforms enhance planning for structural heart and aortic procedures by optimizing device selection and anticipating complications. Intraoperatively, AI augments robotic precision, stabilizes instrument motion, identifies anatomy through computer vision, and predicts hemodynamic instability via real-time waveform analytics. Integration of the Hypotension Prediction Index into perioperative pathways has already demonstrated reductions in ventilation duration and improved hemodynamic control. Postoperatively, machine-learning early-warning systems and physiologic waveform models predict acute kidney injury, low-cardiac-output syndrome, respiratory failure, and sepsis hours before clinical deterioration, while emerging closed-loop control and remote monitoring tools extend individualized management into the recovery phase. Despite these advances, current evidence is limited by retrospective study designs, heterogeneous datasets, variable transparency, and regulatory and workflow barriers. Nonetheless, rapid progress in multimodal foundation models, digital twins, hybrid OR ecosystems, and semi-autonomous robotics signals a transition toward increasingly precise, predictive, and personalized cardiac surgical care. With rigorous validation and thoughtful implementation, AI has the potential to substantially improve safety, decision-making, and outcomes across the entire cardiac surgical continuum. Full article

Three-dimensional (3D) printing has transformed the development of microneedle arrays (MNAs) by enabling exceptional control over their geometry, distribution, materials, and functionality in a single-step, customizable process. This review represents a design-centric framework that organizes recent advancements in four interconnected levers: (i) individual microneedle (MN) geometry and size; (ii) patch-level MN distribution and multi-array architectures; (iii) computer-aided design (CAD), finite element analysis (FEA), computational fluid dynamics (CFD), and artificial intelligence/machine learning (AI/ML)-driven optimization; and (iv) manufacturing constraints and emerging solutions for scalability and reproducibility. Outcomes show that small changes in the radius of the MN’s tip, the MN’s aspect ratio, the MN’s internal lattice architecture, and the spacing of the array can dramatically influence their insertion force, mechanical reliability, payload capacity, and therapeutic coverage. Now, digital tools can bridge the design and experimental outcomes, while novel morphologies, hybrid materials, and theranostic integrations are expanding the clinical potential of MNs. The remaining challenges, resolution-versus-throughput trade-offs, biocompatibility, batch-to-batch consistency, and lack of testing standardization are examined alongside promising directions in high-throughput 3D printing, stimuli-responsive materials, and closed-loop systems. Finally, rational, model-guided design strategies are positioning 3D-printed MNAs as versatile platforms for painless, patient-specific drug delivery, diagnostics, and personalized medicine. Full article

Artificial intelligence (AI) is rapidly transforming anesthesiology practice across perioperative settings. This review explores the evolution and implementation of hybrid decision-making models that integrate AI capabilities with human clinical expertise. From historical foundations to current applications, we examine how machine learning algorithms, deep learning networks, and big data analytics are enhancing anesthetic care. Key applications include perioperative risk prediction, AI-assisted patient education, automated analysis of clinical records, airway management support, predictive hemodynamic monitoring, closed-loop anesthetic delivery systems, and pain management optimization. In procedural contexts, AI demonstrates promising utility in regional anesthesia through anatomical structure identification and needle navigation, monitoring anesthetic depth via EEG analysis, and improving quality control in endoscopic sedation. Educational applications include intelligent simulators for procedural training and academic productivity tools. Despite significant advances, implementation challenges persist, including algorithmic bias, data security concerns, clinical validation requirements, and ethical considerations regarding AI-generated content. The optimal integration model emphasizes a complementary approach where AI augments rather than replaces clinical judgment—combining computational efficiency with the irreplaceable contextual understanding and ethical reasoning of the anesthesiologist. This hybrid paradigm reinforces the anesthesiologist’s leadership role in perioperative care while enhancing safety, precision, and efficiency through technological innovation. As AI integration advances, continued emphasis on algorithmic transparency, rigorous clinical validation, and human oversight remains essential to ensure that these technologies enhance rather than compromise patient-centered anesthetic care. Full article

Background/Objectives: Type 1 diabetes (T1D) is a common childhood condition with rising global incidence. Because early-onset T1D coincides with key periods of brain maturation, affected children may face neurocognitive risks. This review summarizes current evidence on the neurocognitive impact of pediatric T1D and related clinical implications. Methods: A structured search of PubMed, Scopus, and Web of Science (inception–October 2025) used combinations of terms related to T1D, cognitive outcomes, and brain imaging. Studies involving participants under 18 years that reported cognitive or neuroimaging findings were included. Results: Diabetic ketoacidosis (DKA) at diagnosis is consistently linked with acute and longer-term neurological injury, including reduced brain volume and potential persistent deficits in memory and executive functioning. Severe or recurrent hypoglycemia disproportionately affects the hippocampus, contributing to lasting learning and memory impairments. Chronic hyperglycemia is a major driver of progressive neurocognitive decline; higher HbA1c is associated with smaller brain volumes and poorer executive function, attention, and processing speed. Early-onset disease and longer duration further increase vulnerability. These neurocognitive effects translate into modest reductions in academic performance and quality of life, especially with poor glycemic control. Emerging evidence suggests that continuous glucose monitoring, insulin pumps, and hybrid closed-loop systems improve metabolic stability and may support healthier brain development. Conclusions: T1D children experience subtle but meaningful neurocognitive risks shaped by glycemic extremes and early disease onset. Routine neuropsychological monitoring, strengthened academic support, and wider use of advanced diabetes technologies may help preserve cognitive development. Larger, longitudinal neuroimaging studies are needed to guide targeted neuroprotective strategies. Full article

Lignocellulosic residues are increasingly explored as alternatives to wood in particleboard production, fostering sustainability within the circular economy. Beyond these, non-lignocellulosic wastes such as plastics are gaining attention for enhancing panel durability and performance. This study evaluates waste high-density polyethylene (HDPE) as a partial substitute for adhesive resin in sugarcane bagasse-based medium-density particleboards. The objective was to valorize agricultural and plastic residues while reducing reliance on petroleum-based resins and associated environmental impacts. Panels (750 kg/m³) were produced with two face layers of sugarcane bagasse and a core layer combining bagasse and HDPE, bonded with castor oil-based polyurethane resin at 8% and 12% contents. Physical and mechanical performance was assessed against national and international standards, complemented by natural and accelerated weathering tests. A comparative life cycle assessment (LCA) was conducted to benchmark hybrid panels against conventional particleboards. Results showed that incorporating HDPE allows for resin reduction without compromising performance, meeting standard requirements for several applications. The LCA indicated lower environmental burdens in 8 of 10 impact categories for hybrid panels relative to conventional ones, underscoring their potential to reduce fossil resource use and emissions. The findings demonstrate that integrating waste plastics into particleboard production not only improves resource efficiency but also delivers tangible environmental benefits. This approach offers a scalable pathway for advancing sustainable materials, closing waste loops, and supporting circular economy practices in the wood-based panel industry. Full article

Lithium-ion batteries (LIBs) dominate energy storage for electric vehicles (EVs) due to their high energy density, long cycle life, and low self-discharge. However, high costs, complex manufacturing, and the requirement for advanced battery management systems (BMSs) constrain their broader deployment. Therefore, extending the utility of LIBs through reuse is essential for economic and environmental sustainability. Retired EV batteries with 70–80% state-of-health (SOH) can be repurposed in battery energy storage systems (BESSs) to support power grids. Effective reuse depends on accurate and rapid assessment of SOH and state-of-safety (SOS), which relies on precise state-of-charge (SOC) detection, particularly for aged LIBs with elevated thermal and electrochemical risks. This review systematically surveys SOC, SOH, and SOS detection methods for second-life LIBs, covering model-based, data-driven, and hybrid approaches, and highlights strategies for a fast and reliable evaluation. It further examines power electronics topologies and control strategies for integrating second-life LIBs into power grids, focusing on safety, efficiency, and operational performance. Finally, it analyzes key factors within the closed-loop supply chain, particularly reverse logistics, and provides guidance on enhancing adoption and supporting the establishment of circular battery ecosystems. This review serves as a comprehensive resource for researchers, industry stakeholders, and policymakers aiming to optimize second-life utilization of traction LIBs. Full article

High-valent metal species (iron, manganese, cobalt, copper, and ruthenium) based advanced oxidation processes (AOPs) have emerged as sustainable technologies for water remediation. These processes offer high selectivity, electron transfer efficiency, and compatibility with circular chemistry principles compared to conventional systems. This comprehensive review discusses recent advances in the synthesis, stabilization, and catalytic applications of high-valent metals in aqueous environments. This study highlights their dual functionality, not only as conventional oxidants but also as mechanistic mediators within redox cycles that underpin next-generation AOPs. In this review, the formation mechanisms of these species in various oxidant systems are critically evaluated, highlighting the significance of ligand design, supramolecular confinement, and single-atom engineering in enhancing their stability. The integration of high-valent metal-based AOPs into photocatalysis, sonocatalysis, and electrochemical regeneration is explored through a newly proposed classification framework, highlighting their potential in the development of energy efficient hybrid systems. In addition, this work addresses the critical yet underexplored area of environmental fate, elucidating the post-oxidation transformation pathways of high-valent species, with particular attention to their implications for metal recovery and nutrient valorization. This review highlights the potential of high-valent metal-based AOPs as a promising approach for zero wastewater treatment within circular economies. Future frontiers, including bioinspired catalyst design, machine learning-guided optimization, and closed loop reactor engineering, will bridge the gap between laboratory research and real-world applications. Full article

Background: Advanced hybrid closed-loop (aHCL) systems have improved glycemic control in individuals with type 1 diabetes (T1DM). However, it remains unclear whether their efficacy and safety differ by patient’s sex, in view of known sex-related physiological and behavioral differences in disease control and management. Methods: This retrospective, single-center study included 176 adults with T1DM starting aHCL therapy with Medtronic MiniMed™ 780G. Continuous glucose monitoring (CGM) metrics, glycated hemoglobin (HbA1c), and glycemic variability (GV) indexes were collected at baseline, 6 months, and 12 months after starting aHCL therapy. Only patients with at least 70% sensor usage were included at each time point. The primary outcome was the assessment of sex-related differences in CGM metrics at 12 months. Secondary outcomes included changes in HbA1c and GV indexes by sex and over time. Results: TIR increased significantly at 6 (+6.6%, p < 0.001) and 12 months (+5.4%, p < 0.001), TAR decreased, and TBR remained stable. HbA1c was significantly reduced at both 6 and 12 months (−0.6%, p < 0.001). Improvements were consistent in both males and females, with females exhibiting better improvement in HbA1c compared to males (−0.4%, p = 0.049). No significant sex differences were found in CGM metrics at 12 months. GV indexes improved significantly in both groups, regardless of sex. At the multivariable analysis, only HbA1c <7.0% at baseline was associated with the achievement of the composite outcome (TIR > 70%, TBR < 4%, HbA1c < 7.0%). Conclusions: aHCL therapy improved glycemic control and GV in adults with T1DM, regardless of the patient’s sex. These results support the generalizability of aHCL therapy and underscore the need to ensure equitable access to technologies rather than sex-specific adjustments. Full article