Under weak grid scenarios, wide variations of grid impedance distort resonance characteristics of LCL-type grid-connected inverters. Digital control delays introduce phase lag, which easily causes damping polarity reversal in conventional capacitor-current-feedback active damping strategies. From the perspective of impedance stability, this paper reveals that control delays produce frequency-dependent resistive components in equivalent damping impedance. The analytical boundary of positive-negative resistance transition is derived, which dominates the weak-grid adaptability of inverters. Accordingly, an impedance reshaping strategy based on phase-lead delay compensation is proposed. Embedded in the feedback loop, the phase-lead network extends the valid positive-resistance frequency region and decouples the inherent coupling between LCL resonance frequency and sampling frequency. The critical frequency is lifted from [Formula: see text] to above [Formula: see text], and the system maintains a stability margin over 45° within 0-10 mH grid inductance range. A quasi-proportional-resonant cascaded current regulator is further designed to suppress background harmonic interference. Simulation and experimental tests on a 5 kW prototype verify the superior performance. When grid inductance steps from 0 to 8 mH, grid-connected current THD remains below 2.8%, and transient response completes within two fundamental cycles. This study provides theoretical guidance and practical solution for stable grid integration of high-penetration renewable energy systems.
This paper proposes a three-tier Stackelberg game-based hierarchical optimization framework for integrated electric vehicle (EV) battery swapping stations (BSS) and charging point operator (CPO) systems. The framework models the strategic interactions among three decision-making layers comprising grid operators, integrated CPO-BSS operators, and EV users within a multi-stakeholder energy management environment. A bi-level mixed-integer linear programming (MILP) formulation combined with backward-induction-based Subgame Perfect Nash Equilibrium (SPNE) analysis is developed to optimize dynamic electricity pricing, battery charging and swapping schedules, grid power utilization, and user service decisions under operational and grid constraints. The upper layer determines time-varying tariffs, demand-response incentives, and capacity charges to improve grid stability and social welfare, while the middle layer optimizes integrated charging-swapping operations and battery inventory management in response to grid signals and user behavior. The lower layer models EV users as rational followers responding to dynamic pricing through charging or swapping decisions. The proposed framework is validated using EV charging sessions from the publicly available ACN-Data corpus from which BSS swapping demand inputs were synthetically derived via a principled data-mapping procedure and Italian GME day-ahead electricity market price data. The results show that the proposed hierarchical framework reduces the operational cost of the system by 14.2-26.5% when compared with the unoptimized baseline system over the five-year simulation period (2020-2024), while reducing the peak grid demand by 26-28% (192-204 kW) compared with the unoptimized system and maintaining 96.8% service reliability. The coordinated strategy further enables effective load shifting toward low-price periods, enhances battery utilization efficiency, and improves demand elasticity through dynamic pricing mechanisms. Comparative analysis shows that the proposed framework captures 15-22% additional value over decentralized Nash equilibrium strategies while achieving near-optimal centralized social welfare performance under realistic institutional and operational constraints. Sensitivity and benchmarking studies confirm the robustness, computational tractability, and scalability of the proposed approach across varying tariff structures, battery inventories, and demand scenarios. The framework provides practical insights for EV infrastructure planning, grid-aware energy management, and regulatory policy design for future integrated charging and battery swapping ecosystems.
To investigate the structural and vascular changes of central and peripheral retina/choroid in patients with primary aldosteronism (PA) using ultra-widefield (24 × 20 mm²) swept source optical coherence tomography angiography (SS-OCTA), and explore the relationship between systemic factors and ocular indicators. This is a prospective study with both cross-sectional and longitudinal self-controlled comparisons. A total of 61 PA patients (122 eyes) due to unilateral adrenal adenoma and 59 healthy individuals (118 eyes) were recruited. The parameters assessed included retinal/choroidal thickness (RT, ChT), choroidal vascular index (CVI), choroid vessel/stromal volume per unit area (CVV/a, CSV/a), and deep capillary plexus vessel density (DCPVD). SS-OCTA images were divided into 3 × 3 grid, with the grid center defined as central area and the rest as peripheral area. And receiver operating characteristics curves were generated to determine the discriminative power of choroidal parameters. Compared to healthy controls, PA patients exhibited increased RT, ChT, CVI, CSV/a, CVV/a and DCPVD (P < 0.05). After adrenalectomy, the decline in aldosterone levels was accompanied by the recovery of choroidal parameters (P < 0.05). After controlling confounding factors, SFCT and mean peripheral ChT were related to aldosterone and serum potassium, while central and mean peripheral CVI metrics were correlated with serum potassium. The AUC values of choroidal parameters were greater in the combination of peripheral subfields than the central area alone. PA patients demonstrated abnormal retinal/choroidal thickness and choroidal volume. Choroidal parameters were positively correlated with endogenous aldosterone levels, underscoring the pivotal role of aldosterone in ocular health.
The use of artificial intelligence and machine learning (ML) tools is now common in the advancement of health care services and clinical risk estimation. Legacy systems make use of highly informative feature sets developed from years of clinical expertise and research to estimate different outcomes, but only recently have they been tested against novel statistical approaches. One such system, the Johns Hopkins Adjusted Clinical Group (ACG) System, is a long-standing and widely used approach to categorizing clinical risk factors, and it is amenable to ML techniques. This study aims to test the ACG System using a contrasted area under the receiver operating characteristic (AUROC) and F1 classification optimization strategy and compare its performance against traditional logistic regression methods. Assuming that selected ML algorithms can be tuned to enhance overall measures of performance, this would strengthen arguments for incorporating them into ACG-related workflows. Using a retrospective observational design, prospective year estimates of all-cause hospitalization and elevated total cost were modeled using a cross-validation framework. Patients with elevated costs were identified as those falling above the 95th percentile of total amounts billed, including pharmacy costs. Hyperparameter settings for XGBoost (Extreme Gradient Boosting), random forest, and elastic net were determined using average cross-validated performances for F1 and AUROC in a grid search aimed at maximizing either statistic. Additional iterated cross-validation was used to compare point-estimated average AUROC and F1-scores between models, further decomposed by sensitivity, positive predictive value, and F-beta statistics. There were 350,463 patients selected in 2019 from the Johns Hopkins Health System. Model features identified by the ACG System for predicting prospective year hospitalization and total cost were included in these analyses. Findings suggest small but statistically significant improvements in cross-validated AUROC and F1-scores over logistic regression, using either optimization strategy and XGBoost. Logistic models achieved average receiver operating characteristic values of 0.886 and 0.841 for cost and hospitalization, respectively, whereas XGBoost achieved 0.891 and 0.849, respectively. F1 optimization yielded similar findings, with logistic models achieving 0.367 and 0.341 on average for hospitalization and cost, respectively, but XGBoost exceeded values for cost but not for hospitalization (0.411 and 0.328, respectively). The clinical implications of these findings and the effect of class imbalance on model calibration are explored, along with the limitations of these data and approach. The core finding is that logistic regression remains very well-suited to these tasks, especially in situations where the efficiency or interpretability of models is critical. Under conditions of imbalance, regressions tended to yield high-precision estimates for the outnumbered class. Nevertheless, the findings also underscore a diversity of suitable models depending on clinical use cases, each having its own tradeoffs for evaluating performance. As such, health systems must clearly identify the needs and expectations of a model before calibrating one for use.
Low back pain affects hundreds of millions of people and is associated with degenerative changes in the lumbar spine. Trabecular bone microarchitecture change is a reasonable contributor to low back pain, but it remains largely invisible on routine clinical CT because typical voxel sizes (500 to 1000 μm) are insufficient to resolve trabeculae (about 100-200 μm). We present LumbarSR, a paired and registered dataset of 30 human lumbar vertebral specimens scanned with a photon-counting micro-CT (Micro-PCCT) reference at 105 μm isotropic resolution and with a standard clinical CT system under eight acquisition configurations formed by the factorial combination of two in-plane resolutions (195 and 586 μm), two slice thicknesses (500 μm and 1000 μm), and two reconstruction kernels (bone and soft tissue). Clinical CT volumes are rigidly registered to the Micro-PCCT reference using ANTs-based alignment and resampled to a common voxel grid, enabling voxel-wise evaluation and supervised learning. LumbarSR provides data in both original DICOM and registered NIfTI volumes with a consistent directory structure and specimen identifiers. We provide baseline evaluations using whole-image and masked image-quality metrics, trabecular morphometry against the Micro-PCCT reference, and super-resolution benchmarks based on interpolation and deep learning methods. LumbarSR is intended to support the development and evaluation of super-resolution methods for lumbar vertebra CT and related analyses of trabecular level structure. Because specimens were de-identified dry teaching specimens without available clinical histories or demographic metadata, LumbarSR should be interpreted as a paired imaging and benchmarking resource rather than a clinically labeled cohort.
To understand the spatiotemporal variations of vegetation greenness and the differences in its response to climate and human activities based on land cover change in subtropical region, this paper focused on the spatiotemporal characteristics of vegetation greenness (indicated by NDVI) and its sensitivity to temperature, precipitation and human activities (indicated by nighttime light (NTL)) under different land cover scenarios in Guangdong, China. The contributions of climate and human activities under land cover change to the variations in vegetation greenness were quantified using Lindeman, Merenda and Gold method (LMG). The conclusions revealed that (1) NDVI was significantly higher in regions with land cover change. This might be because the conversion from grassland to evergreen forest increased vegetation density. (2) NDVI in regions with unchanged land cover exhibited higher correlations with temperature and NTL. At the grid scale, NDVI was more sensitive to temperature rather than precipitation and was significantly correlated with NTL in most regions. (3) At the province scale, LMG multivariate regression model indicated that land cover change was the dominant factor affecting vegetation greenness with a contribution of 54.2%, while the contribution of climate factors was relatively low with temperature contributing 18.2% which was greater than the contribution of precipitation (2.2%). This paper contributes to a deeper understanding of the mechanisms underlying vegetation responses to the combined effects of climate change and human activities, providing a scientific basis for regional ecological quality assessment and ecosystem management.
The increasing demand for low-carbon construction materials has led to growing interest in carbonation-cured concrete, which enables permanent sequestration of carbon dioxide while influencing material performance. However, predicting the compressive strength of CO₂-cured concrete remains challenging due to complex nonlinear interactions between mixture composition and curing conditions. Unlike previous studies that employed single algorithms with grid search optimization, this study integrates seven advanced ensemble models with metaheuristic optimization, multi-level interpretability analysis, and probabilistic uncertainty estimation within a unified framework. Seven ensemble learning models, including Random Forest, CatBoost, LightGBM, NGBoost, AdaBoost, Extra Trees, and DeepGBM, were optimized using Particle Swarm Optimization to improve predictive performance. The Random Forest model achieved the highest predictive accuracy on the testing dataset with R2 = 0.955 and RMSE = 3.9251 MPa. All seven models demonstrated strong predictive performance, with R2 values exceeding 0.92, confirming the effectiveness of PSO-based hyperparameter optimization across all evaluated algorithms. SHapley Additive exPlanations, accumulated local effects, permutation feature importance, and counterfactual analysis were applied to evaluate feature contributions and interactions. Cement content, coarse aggregate content, and water content were identified as the most influential variables, while carbonation curing parameters showed secondary effects. A graphical user interface was developed to enable real-time compressive strength prediction based on user-defined input parameters. The proposed framework provides accurate prediction and interpretable insights into the influence of mixture composition and curing conditions within the scope of the studied dataset.
In recent years, the field of medical imaging has witnessed substantial progress due to the integration of advanced machine learning techniques, particularly in the diagnosis of critical conditions such as breast cancer. This study aims to improve the predictive accuracy of breast cancer diagnosis using ultrasound images by employing the cross-attention multi-scale vision transformer (CrossViT). The proposed methodology involves a dual-branch architecture in which each branch processes image patches of different sizes, thereby capturing both fine-grained and coarse-grained features. The model incorporates a cross-attention mechanism that efficiently fuses these multi-scale features, enhancing its ability to discern complex patterns in medical images. A public ultrasound dataset was partitioned at the patient level using a stratified 80/10/10 train/validation/test split. The development data used for model optimization included 4074 benign and 4042 malignant training images, along with 500 benign and 400 malignant validation images after preprocessing and augmentation, and final model performance was assessed on a held-out test set. Hyperparameters were fine-tuned using a grid search strategy to optimize performance, and training was conducted with stochastic gradient descent and regularization techniques to support stable convergence. Results from this single-dataset experiment showed that CrossViT yielded higher observed performance metrics than the evaluated ResNet architectures across accuracy, precision, recall, F1-score, and AUC. However, these findings should be interpreted as exploratory comparative observations rather than statistically confirmed evidence of model superiority. In conclusion, CrossViT represents a promising technical approach in medical imaging and may have potential utility for automated breast cancer diagnosis in future clinically validated settings.
Solar radiation forecasting is a complex task since the radiation signal is nonlinear, intermittent and is significantly influenced by meteorological variability, which makes it vital for PV planning, renewable energy planning and stability of the smart grid. In this work, a replicable comparison between CNN-LSTM and CNN-BiLSTM models for one-step ahead solar clearness-index forecasting based on multivariate climate variables from NASA POWER dataset for Delhi, India, is presented. Under identical preprocessing, windowing, chronological splitting, and training conditions, CNN-LSTM achieved MAE = 0.0880, RMSE = 0.1100, R2 = 0.3100, EVS = 0.3154, WI = 0.6317, and APB = 1.89%, whereas CNN-BiLSTM obtained MAE = 0.1015, RMSE = 0.1224, R2 = 0.1456, EVS = 0.1998, WI = 0.5261, and APB = 5.98%. The Skill Scores shown and the negative values for direct clearness-index prediction do not imply that the persistence reference was unattainable, but rather reveal that the results are a controlled model-to-model comparison and not evidence of state-of-the-art superiority. Reconstructed all-sky irradiance produced stronger agreement with observations (MAE = 0.4353, RMSE = 0.5417, R2 = 0.7884, EVS = 0.7965, WI = 0.9299, and APB = 3.90%). The main task of CNN-LSTM is to provide a practical balance between accuracy and efficiency in this experimental context, and further testing with other locations, more powerful baselines and probabilistic forecasting techniques is needed.
Computational modeling offers a principled way to link the structure of semantic networks to the processes that support creative problem solving. We combined item-level semantic network analyses (Experiment 1, behavioral method) with the Spreading Activation Model simulation (SAM; Experiment 2, computational modeling) for the Chinese Word Remote Associates Test (CWRAT). In Experiment 1, we built customized item-semantic networks based on the CWRAT-free association task and quantified network properties for each item. Results showed that greater network efficiency and higher modularity were associated with higher item pass rates, whereas larger network size was associated with lower item pass rates. In Experiment 2, we used SAM to simulate activation dynamics on each item's semantic network across a grid of retention (r, the proportion of activation retained at each time step) and decay (d, the proportion of activation lost at each time step) parameters. At the item level, we extracted target-activation metrics (mean maximum activation, mean activation, and mean activation retention duration), and the results showed that these metrics correlated positively with the empirical item pass rate and negatively with item response time. At the individual level, we estimated best-fitting parameter pairs (r, d) for each participant. Results showed that these parameters correlated negatively with individuals' performance in both full and high-difficulty item sets. Particularly on the high-difficulty subset, results showed that parameters correlated positively with individual ratings of insightfulness and interest. Together, these results describe how bottom-up spreading activation may interact with top-down control to shape and support the creative thinking process.
To compare target coverage, dose gradient, organ-at-risk (OAR) sparing, and low-dose bath between CyberKnife (CK) and noncoplanar linac-based techniques, including noncoplanar volumetric modulated arc therapy (VMAT) and noncoplanar three-dimensional conformal radiotherapy (3DCRT), for left-sided accelerated partial breast irradiation (APBI). Fifteen patients previously treated with CyberKnife for left-sided APBI were retrospectively replanned using noncoplanar VMAT and noncoplanar 3DCRT on the Monaco treatment planning system. All plans were generated using a 2-mm dose calculation grid and normalized to achieve comparable target coverage (V98% ≥ 95% of the prescribed dose of 30 Gy in 5 fractions). Dose-volume parameters for planning target volume (PTV), OARs, gradient index (GI), conformity index (CI), and low-dose bath volumes were analyzed. Statistical comparison was performed using paired two-tailed t-tests. CyberKnife demonstrated significantly superior dose conformity and gradient, with the lowest GI (2.65) compared with VMAT (3.22) and 3DCRT (3.52) (p < 0.01). Mean doses and low-dose volumes (V3 to V20 Gy) to the heart, left anterior descending (LAD) coronary artery, and left ventricle (LV), ipsilateral lung, contralateral lung, contralateral breast, and chest wall were significantly lower with CyberKnife. However, the use of noncoplanar beam arrangements in VMAT and 3DCRT resulted in clinically acceptable dose fall-off and comparable GI values, particularly between noncoplanar VMAT and 3DCRT. Target coverage metrics (V98% and V100%) were comparable across all techniques. CyberKnife provides superior dose gradient and reduced low-dose bath for left-sided APBI. Nevertheless, noncoplanar VMAT and noncoplanar 3DCRT demonstrate favorable and clinically acceptable dosimetric performance, indicating that noncoplanar linac-based techniques may serve as viable alternatives where CyberKnife is unavailable.
Effective communication during laparoscopic procedures is frequently undermined by spatial disorientation and inconsistent terminology between instructors and trainees. This study examined whether standardized visual overlays on endoscopic monitors could enhance communication and learning. We conducted a three-phase mixed-methods study: qualitative observation of 20 laparoscopic teaching cases; a randomized trial of 63 second-year medical students assigned to control, clock, or alphanumeric grid (AG) overlays during three trials of a standardized transfer task; and intraoperative implementation in 44 cases (30 AG, 14 clock) with post-case surveys and qualitative feedback. In simulation, the clock overlay produced the fastest completion times, whereas the AG yielded the lowest error scores, and both overlays outperformed the control. Intraoperatively, the AG was rated higher than the clock for communication clarity, spatial orientation, perceived operative efficiency, and trainee confidence. Standardized visual overlays, particularly the AG, appear to support intraoperative teaching by providing a shared spatial frame of reference.
Chronic ankle instability (CAI) is common in athletic populations and is associated with recurrent sprains and impaired performance. Recent evidence shows that fibularis longus has distinct anterior and posterior regions and that, during isometric eversion, people with CAI display a reduced relative contribution of the posterior region. This muscle is typically trained with band-resisted eversion in supine (BES) and BOSU-based exercises (seated and single-leg eversion, BOSU-Sit and BOSU-SL, respectively), yet it is unknown whether these specific tasks correct or perpetuate regional activation deficits. As exercise demands increased from BES to BOSU-based tasks, fibularis longus activation would be differentially modulated and people with CAI would exhibit a greater anterior shift of the activation barycenter together with a reduced relative contribution of the posterior region, particularly during the BOSU-based tasks. Descriptive laboratory study. Level 5. A total of 40 physically active adults (CAI, n = 20; no-CAI, n = 20) performed 3 exercises: BES, BOSU-Sit, and BOSU-SL. High-density surface electromyography recorded fibularis longus activation using a 64-electrode grid. Root mean square was computed for anterior and posterior regions, and the activation barycenter was calculated along the x-axis (anteroposterior) and y-axis (cephalocaudal). In CAI, both BOSU exercises shifted the x-axis barycenter anteriorly relative to BES (BOSU-Sit, P = 0.01; BOSU-SL, P = 0.01). BES was the only exercise in which people with CAI exhibited greater posterior-region activation than controls (P = 0.003). In contrast, BOSU-SL produced an anterior-dominant pattern, with greater anterior-region activation than the posterior region (P = 0.04) and lower posterior-region activation than BES (P = 0.008). Therapeutic exercises commonly grouped as "fibular strengthening" are not neuromuscularly equivalent. In this acute task comparison, BES was associated with a more posterior barycenter and greater posterior-region activation in CAI, whereas BOSU-based tasks primarily activated the anterior region.
Sodium polysulfide nonaqueous redox-flow batteries are promising candidates for grid-scale energy storage due to their high theoretical energy density and earth-abundant components. However, their performance is fundamentally limited by severe polysulfide shuttling and unstable sodium-metal interfaces, particularly under high-concentration catholyte conditions. Here, we report a scalable and perfluoroalkyl-substances-free membrane modification strategy via a simple dip-coating method to fabricate commercial glass fiber (GF) with poly(4-styrenesulfonic acid-co-maleic acid) sodium salt (PSSMA). The sulfonate and carboxylate functional groups in PSSMA provide both electrostatic and steric repulsion against polysulfide anions, while preserving the mechanical flexibility of the GF substrate. The optimized PSSMA-GF membrane significantly suppresses shuttle current, mitigates self-discharge, and delays the formation of short-chain, less soluble polysulfide species, leading to improved Coulombic efficiency (from <60 to 98.3%) and stable cycling over 100 cycles in sodium polysulfide coin-type full cells. Furthermore, the coating promotes uniform solid electrolyte interphase formation on the sodium anode. The feasibility of the PSSMA-GF membrane was further demonstrated using the flow-cell configuration, showing a smooth discharge plateau at 1.5 V under 1.0 mA cm-2. This work demonstrates an environmentally benign, cost-effective, and easily scalable membrane fabrication strategy, offering a practical pathway to overcome key challenges in sodium polysulfide redox-flow batteries and other advanced energy storage systems.
The rapid urbanization has posed a serious threat to the ecological security of river basins. Exploring the trade-offs, synergies, driving factors, and ecological management zoning of regional ecosystem services is of great significance for achieving the sustainable utilization of ecosystems in the Yangtze River Basin. We employed the InVEST model to quantitatively assess the spatial patterns of five major ecosystem services (soil conservation, water yield, habitat quality, and carbon sequestration and food supply) from 2000 to 2023, and used the Spearman correlation coefficient to examine the trade-off and synergy relationships among different ecosystem services and their scale effects. We then applied the Geodetector model and random forest method to identify the dominant driving factors, interactive effects and nonlinear response characteristics, while used the K-means clustering method to identify ecosystem service bundles. Based on these analyses, we proposed ecological management zones and their optimization pathways for the Yangtze River Basin. The results showed that carbon sequestration, habitat quality, and soil conservation exhibited a slight declining trend from 2000 to 2023 by 0.5%, 2.0%, and 10.2%, respectively, whereas water yield and food supply increased by 1.0% and 14.6%. The trade-offs and synergies among ecosystem services displayed significant scale effects. As the spatial scale expanded from grid to county and city levels, the weak synergistic relationships gradually strengthened and became dominant, while the trade-off intensity weakened and the spatial distribution became more clustered. The key driving factors of the five ecosystem services varied significantly. There was also a distinct nonlinear response and a threshold effect: vegetation coverage (explanatory power q=0.45) dominated carbon sequestration, population density (q=0.71) mainly affected habitat quality, precipitation determined (q=0.93) water yield, and slope (q=0.60) affected soil conservation, population density (q=0.56) affected food supply. Based on the clustering of ecosystem services, we classified the study area into three ecological management zones, namely ecological protection zones, ecological conservation zones, and provisioning service zones. We proposed differentiated optimization strategies for each zone. 快速的城镇化进程对流域的生态安全造成威胁,探索区域生态系统服务的权衡协同关系、驱动因素及生态管理分区,对实现长江流域生态系统的可持续利用具有重要意义。本研究通过InVEST模型对2000—2023年间长江流域的土壤保持、产水量、生境质量、固碳量和粮食供给这5种主要生态系统服务进行空间定量化测算,运用Spearman相关系数探究不同生态系统服务之间的权衡协同关系及其尺度效应,并采用地理探测器和随机森林方法探究其主要影响因素、交互作用与非线性响应特征,进而结合K-Means聚类分析法识别生态系统服务簇,探索长江流域的生态管理分区及其优化路径。结果表明:2000—2023年间,长江流域的固碳量、生境质量、土壤保持均小幅下降,分别下降了0.5%、2.0%、10.2%,产水量和粮食供给分别上升了1.0%和14.6%。长江流域生态系统服务的权衡协同关系具有显著的尺度效应,并随着空间尺度从栅格扩大到县域和市域,其弱协同关系逐渐加强成为主导,而权衡强度减弱且空间分布更加集中。长江流域各生态系统服务的关键驱动因子存在显著差异,且存在明显的非线性响应和阈值效应。其中,植被覆盖率(解释力q=0.45)主导固碳量,人口密度(q=0.71)主要影响生境质量,降水(q=0.93)决定产水量,坡度(q=0.6)决定土壤保持,人口密度(q=0.56)决定粮食供给。基于生态系统服务簇,将研究区划分为3个生态管理分区,分别为生态防护区、生态保育区及供给服务区,并提出了差异化的分区优化路径。.
The North American swine industry is shifting towards raising pigs to heavier weights, partly due to the potential for higher profitability for pork processors. However, producers may incur losses due to feed and transportation costs associated with heavy-weight pigs. In some cases, these animals can fall outside grid-based grading specifications, resulting in discounts for carcasses that are too heavy or too lean/too fat. Furthermore, increasing weight raises concerns regarding nutritional requirements, feed efficiency, environmental sustainability, welfare issues, carcass composition, chilling rate, meat quality, and implications for genomic selection. Variation exists for growth performance traits and carcass composition across genotypes. In this review, we summarise the biological, physiological, and genetic mechanisms underlying the effect of increasing weight in contemporary pig production. A comprehensive understanding of animal performance, animal welfare, carcass traits and genetic regulations in heavy-weight pigs will be significant for refining breeding goals, precision feeding strategies and handling practices.
Aqueous zinc-iodine batteries (AZIBs) are emerging as highly promising candidates for next-generation, grid-scale energy storage due to the intrinsic safety of water-based electrolytes, the high theoretical capacity of the zinc anode, and the rapid conversion kinetics of the iodine cathode. However, the practical commercialization of AZIBs is severely impeded by formidable interfacial instabilities, including the uncontrollable growth of zinc dendrites, parasitic hydrogen evolution reactions (HER), and the notorious polyiodide (I3-, I5-) shuttle effect. These macroscopic degradation modes are fundamentally rooted in the robust [Zn(H2O)6]2+ primary solvation sheath and the immense thermodynamic driving force for polyiodide dissolution in highly polar aqueous media. To address these interconnected challenges, electrolyte engineering has evolved into the most potent, holistic strategy. This comprehensive review systematically evaluates the latest advancements in electrolyte engineering for AZIBs. We first deeply decipher the fundamental thermodynamic mechanisms governing Zn2+ desolvation and iodine multiphase conversion. Subsequently, we critically analyze cutting-edge regulation paradigms, including water-in-salt (WIS) and localized high-concentration electrolytes (LHCE), cosolvent networks, functional molecular additives, deep eutectic solvents (DES), and quasi-solid-state hydrogels. By integrating in situ/operando spectroscopic characterizations with multiscale theoretical computations (such as MD and DFT), we elucidate the structure-activity relationships at the atomic level. Finally, we provide strategic perspectives on the future trajectories of the field, emphasizing the stabilization of multi-electron (I-/I0/I+) halogen chemistry, AI-driven high-throughput screening, and the rigorous standardization of Ah-level pouch cell engineering for extreme-environment applications.
In pediatrics, a personalized educational approach is required to reduce medication administration mistakes made by parents, responsible for 84.4% of outpatient medication errors. The Pediatric Instructive Kit for Adherence optimization to MEDication (PIKAMED) tool aims to optimize medication intake by providing clear instructions on administering liquid oral antibiotics (LOAs).The primary objective was to evaluate the impact of PIKAMED on parents of children in complying with a fictitious LOA prescription, in two simulated situations. Secondary objectives were to detect any misunderstandings caused by PIKAMED and optimize the tool based on the parents' experience. A prospective, single center, comparative study was conducted. Participants were parents of hospitalized children aged 28 days to 6 years old, who had never received any particular training on medical care for their children. Parents were sequentially allocated to one of two groups: a control group following the usual practice, and an experimental group using PIKAMED. The steps involved in reconstituting, administering, and storing LOAs were simulated. A specific audit grid was used to assess parents' adherence to the fictitious prescription. Parents' ability to correctly comply with a fictitious prescription for amoxicillin oral suspension was better with PIKAMED. The median overall success rate reached 93% in the PIKAMED group, compared to 54% in the control group (P < .001). PIKAMED was significantly associated with greater success in reconstituting and administering the medication (OR = 42 [3.49; + ∞ [, P = .007). PIKAMED shows significant potential for enhancing the accuracy of dosing pediatric medication, thus reducing medication errors.
Fabric reinforced cementitious matrix (FRCM) composites are progressively applied to strengthen reinforced concrete (RC) members because they offer improved compatibility, durability and fire resistance compared with epoxy-bonded fibre reinforced polymer (FRP) systems. However, predicting the ultimate flexural capacity of FRCM-strengthened beams remains challenging due to the complex interaction of material, geometric and loading parameters. Here, we develop and evaluate ensemble machine learning (ML) models to predict the ultimate load-carrying capacity of FRCM-strengthened RC beams. A curated database of 244 experimental tests from the literature was compiled, comprising 16 input features describing concrete compressive strength, reinforcement ratios, FRCM properties and specimen geometry. After systematic preprocessing and feature encoding, several ensemble regressors were benchmarked and three models Extra Trees, XGBoost and AdaBoost were analysed by 5 fold cross-validation with grid search for hyperparameter optimization. Among them, the Extra Trees regressor achieved the best performance, with a coefficient of determination R² = 0.898, root mean square error RMSE = 21.5 kN and mean absolute error MAE = 11.7 kN. Feature-importance analysis and SHAP-based explainability consistently identified concrete compressive strength, reinforcement ratio and cracking load as dominant predictors, in agreement with structural mechanics. These results demonstrate that ensemble ML models provide a reliable and interpretable tool for assessing the flexural capacity of FRCM-strengthened RC beams, also highlight the potential of data-driven methods to complement empirical and code-based design approaches in structural strengthening.
The shift towards electric vehicle (EV) adoption requires a solid and efficient charging infrastructure. A crucial part of these systems is the DC-DC converter, which connects the power grid or energy storage with the vehicle's battery pack. Conventional single-module DC-DC converters have inherent limitations in addressing the rising demands for higher power levels, enhanced efficiency, and greater operational reliability in current EV charging applications. The interconnection of several DC-DC converter modules is a promising method to overcome these obstacles, enabling increased power output and better fault tolerance. This research explores the performance of a dual-stage bidirectional EV charger featuring different interconnections. A comprehensive small signal averaged model of the DC stage of the dual-stage EV charger is presented for both the single converter and two converters connected in an interleaved configuration, including a stability analysis. Further, a relative assessment of the charger's performance with the various interconnections is performed, and the results are detailed in terms of various performance metrics, including efficiency, ripple percentage, and source power factor. The conclusions obtained from the comparative analysis are explicitly illustrated to facilitate the selection of the appropriate interconnection for a given application.