Under extreme operating conditions, the stability of dry gas seals (DGS) can be disrupted, leading to face contact and excessive leakage. Therefore, optimizing sealing performance becomes critical. Conventional optimization metrics are typically computed based on a fixed film thickness, which lacks physical consistency. This study proposes a novel metric—Stability Maintenance Energy—to quantify the seal's capacity to withstand external disturbances under force-equilibrium conditions. This metric is coupled with the minimum leakage rate to form a set of conflicting optimization objectives. A T-groove DGS is selected as the case study, with optimization performed using the NSGA-II algorithm. To improve computational efficiency, a Kriging surrogate model is employed in place of the high-fidelity thermo–fluid–solid coupling model, with dynamic updates incorporated during the optimization process. Considering manufacturability, the optimal groove configuration is selected from the Pareto front.

Existing methods to summarize posterior inference for mixture models focus on identifying a point estimate of the implied random partition for clustering and validating it using clustering-based loss functions (Wade and Ghahramani, 2018; Dahl et al., 2022). We propose a novel approach for summarizing posterior inference in nonparametric Bayesian mixture models, prioritizing estimation of the mixing measure as an inference target. Moreover, we propose to validate the estimation of the partition via a new perspective which is based on the implied density estimate. One of the key features is the model-agnostic nature of the approach, which remains valid under arbitrarily complex dependence structures in the underlying sampling model. Using a decision-theoretic framework, the proposed methods identify a point estimate using loss functions defined as discrepancies between mixing measures. Estimating the mixing measure then implies inference on the mixture density and the random partition. To define a discrepancy between mixing measures we exploit the discrete nature of the mixing measure and use a version of sliced Wasserstein distance. We introduce two variants for Gaussian mixtures. The first, mixed sliced Wasserstein, applies generalized geodesic projections on a product of Euclidean space and the manifold of symmetric positive definite matrices. The second, sliced mixture Wasserstein, leverages the linearity of Gaussian mixture measures to define a projection.

Cage subsidence following anterior cervical discectomy and fusion (ACDF) is linked with poor bone quality. MRI-derived bone quality scores have been shown to provide valuable insights into postoperative complication risk; however, the optimal MRI-based metric for predicting cage subsidence remains unclear. This study aims to compare the predictive value of different MRI-derived bone quality measures for cage subsidence following ACDF. Patients undergoing single-level ACDF between October 2012 and September 2022 at our institution with at least 6 months of radiographic follow-up were retrospectively evaluated. T1 preoperative MRI scans were used to measure mean, median, and segmental vertebral bone quality (VBQ) scores, and upper, lower, and average endplate bone quality (EBQ) scores. Postoperative and follow-up X-rays were used to identify cage subsidence. Fifty-six patients met the inclusion criteria; 26 developed cage subsidence and 30 did not. Age, sex, surgical indication, cage type, and clinical setting were similar between groups. Mean disc space loss was significantly greater in the subsidence group (3.99 mm vs. 0.37 mm; p < 0.001). All bone quality scores were significantly higher in the subsidence group across all metrics. Mean VBQ (OR = 14.22), segmental VBQ (OR = 8.23), and lower EBQ (OR = 5.54) were strong predictors of subsidence (p < 0.001). ROC analysis showed excellent discrimination for mean VBQ (AUC = 0.821), segmental VBQ (AUC = 0.817), and median VBQ (AUC = 0.817). Interobserver reliability was high for all bone quality metrics (ICC 0.836-0.925). MRI-derived bone quality metrics, particularly VBQ and lower endplate EBQ scores, are strong predictors of cage subsidence following single-level ACDF. These findings reinforce the clinical utility of preoperative MRI as a non-invasive, radiation-free tool for assessing vertebral bone integrity. Incorporating VBQ and EBQ assessments into surgical planning may enhance risk stratification and optimize postoperative outcomes in patients undergoing cervical fusion.

This study investigates an inlet design for a gravitational vortex turbine (GVT), drawing inspiration from the ancient Nazca puquios. The puquios are ingenious subterranean aqueducts constructed by the Nazca culture (c. 100 BC–800 AD) in southern Peru, featuring spiral ojos de agua (water eyes) used to access groundwater and stabilize flow.The primary objective was to enhance vortex stability and overall GVT efficiency under low-head, low-flow operating conditions. A parametric Nazca-type inlet feeding a conical basin was defined by two controlling factors: the number of turns (N) and the inclination angle (θ). The optimal geometry was determined through a 32 full factorial design, computational fluid dynamics (CFD) simulations, and response surface methodology (RSM), with vortex circulation (Γ) serving as the optimization metric. The best-performing inlet configuration (N=4, θ=13∘) yielded Γ=1.3459 m2/s. This circulation level is comparable to that reported for optimized conventional wrap-around inlets at similar flow rates, but uniquely produced a broader and more symmetric vortex structure. Subsequently, two four-bladed runners (one with twisted blades and one with curved cross-flow blades) were evaluated numerically and experimentally using a laboratory-scale prototype operated at a consistent flow rate (Q≈0.00143 m3/s). CFD predicted maximum efficiencies of 15.37% and 17.07% for the twisted and curved runners, respectively, while experimental tests achieved 8.70% and 11.61%, demonstrating similar efficiency (η) versus angular velocity (ω) characteristics. These results indicate reduced hydraulic effectiveness of the Nazca-inspired geometry for the GVT, with experimental efficiencies below those reported in the literature.

Context. Most of the intracyclic variability in the large-scale solar magnetic field comes from the equatorial dipole component of the solar magnetic field. The equatorial dipole component is highly sensitive to the longitude distribution of the active regions. Aims. We quantify the effect of individual active regions on the large-scale solar magnetic field of the solar cycle 24. We study the effect of the longitude distribution of active regions on the strength of the large-scale dipole component. Methods. We used a surface flux transport (SFT) model to simulate the evolution of individual active regions and quantified their effect on the large-scale magnetic field using the recently developed vector sum method. We took advantage of the longitudinal translational invariance of the SFT model and compared the observed solar cycle 24 to the 10 000 simulations of the solar cycle 24 using randomized longitudinal source locations, but otherwise identical flux emergence. Results. We find that taking into account both the axial and equatorial components of the vector sum characterizing the global solar magnetic field sets better constraints on the parameter space of the SFT model than, for example, using the axial dipole moment alone as an optimization metric. We studied the maximum of cycle 24 and identified the recurrent and localized flux emergence in the southern hemisphere as the main culprit behind the rapid strengthening of the large-scale magnetic field in late 2014. We find that during the declining phase of the solar cycle, the strength of the large-scale magnetic field stayed above the median level of randomized simulations ( p &lt; 0.027) for 42 subsequent rotations (from September 2014 to November 2017). This indicates that the longitudinal distribution of active regions is not random and, rather, that it demonstrates a tendency for some regions to emerge at longitudes where their equatorial components reinforce the large-scale equatorial field.

Association of life's essential 8 with cardiovascular outcomes and mortality in adults with prediabetes: mediating role of inflammatory biomarkers.

The hyperspectral image (HSI) is rich in spectral information and has important applications in the field of ground objects classification. However, HSI data have high dimensions and variable spatial–spectral features, which make it difficult for some models to adequately extract the effective features. Recent studies have shown that fusing spatial and spectral features can significantly improve accuracy by exploiting multi-dimensional correlations. Based on this, this article proposes a spectral integration and focused attention network (SIFANet) with a two-branch structure. SIFANet captures the local spatial features and global spectral dependencies through the parallel-designed spatial feature extractor (SFE) and spectral sequence Transformer (SST), respectively. A cross-module attention fusion (CMAF) mechanism dynamically integrates features from both branches before final classification. Experiments on the Salinas dataset and Xiong’an hyperspectral dataset show that the overall accuracy on these two datasets is 99.89% and 99.79%, which is higher than the other models compared. The proposed method also had the lowest standard deviation of category accuracy and optimal computational efficiency metrics, demonstrating robust spatial–spectral feature integration for improved classification.

Computational protein design using machine learning models has advanced rapidly since the introduction of AlphaFold2. There is now a suite of tools that enable in silico design of proteins with desired structures and properties. Most design workflows require fitting a designed backbone with a sequence that stabilizes it, and many machine learning sequence design models have been proposed. These models are trained to recover the native sequence paired with a known structure, a task known as native sequence recovery (NSR). Here, we demonstrate the limitations of optimizing a sequence design model only for NSR. We show that NSR is often misaligned with more important metrics of model performance: the compatibility of the generated sequence with the desired fold and the ability of the model to predict the energetic effects of mutations. We introduce PottsMPNN, which is trained to generate a Potts energy function consisting of single-residue and residue-pair terms from a protein backbone, and we demonstrate that learning a Potts model reduces NSR but improves sequence generation and energy prediction. To further show that NSR is not the optimal metric, we trained PottsMPNN with noised backbone structures and multiple sequence alignments. In tests on held-out data, NSR decreased, but the quality of the designed sequences and energy predictions improved. By demonstrating the limitations of optimizing for NSR and the effectiveness of alternative strategies for avoiding over-optimizing for NSR, our work provides a new direction for the sequence design field.

ABSTRACT Existing methods for homogeneous pixel selection often fail to balance omission (Type I) and commission (Type II) errors. While the Hypothesis Test of Confidence Interval (HTCI) is computationally efficient, its accuracy is compromised by sensitivity to initial estimates. This study introduces Iter-HTCI, an algorithm that enhances selection accuracy through iterative optimization of the reference estimate using Gamma distribution confidence intervals. Simulations and experiments demonstrate that Iter-HTCI achieves statistical power approaching theoretical maxima. Compared to the Baumgartner-Weiss-Schindler (BWS) and HTCI benchmarks, Iter-HTCI improves small-sample accuracy by 74.36% and 43.41%, respectively. Furthermore, it effectively mitigates over-rejection issues in weak scattering zones. Phase optimization metrics show significant enhancement, with Phase Standard Deviation and Sum of Phase Differences reduced by 31.46% and 58.09% compared to HTCI. Although processing time increases slightly relative to HTCI, Iter-HTCI remains significantly more efficient than BWS and delivers superior phase information for time-series deformation retrieval.

Lithium-ion batteries (LIBs) are currently widely used in the electric vehicle sector and have become one of the core components of new energy vehicles. To ensure that the maximum temperature (Tmax) and maximum temperature difference (∆Tmax) remain within acceptable limits after high-rate discharge, this study proposes a novel air-cooled battery thermal management system (BTMS). This BTMS features innovative design elements in its novel battery case walls and contour-following fin structure. Through physical testing of 21,700 LIB discharges and comparative numerical simulations, the accuracy of the simulation model is ensured. Orthogonal experimental analysis is conducted at four distinct levels for each of the four structural factors involved. The final results demonstrate that the novel battery pack wall and contour-shaped fin structure proposed in this paper significantly enhance the heat dissipation capability of air-cooled BTMS. The proposed Model 9 configuration exhibits optimal thermal performance metrics. The Tmax after 3C rate discharge reaches 39.4 °C, with a ∆Tmax of 7.4 °C. This study demonstrates significant application potential in the structural implementation of air-cooled BTMSs.

Enterprise environments across financial services, insurance operations, and climate risk assessment demand analytical platforms with continuous operational requirements while processing large-scale geospatial and exposure datasets with real-time delivery requirements. Production-grade risk platforms differ fundamentally from experimental systems in their direct responsibility for underwriting decisions, capital allocation strategies, regulatory compliance reporting, and disaster preparedness protocols. Failures or analytical inaccuracies in platforms engender severe economic disruption and adverse societal outcomes. This article investigates large-scale modernization initiatives that deploy cloud-native, artificial intelligence-enabled risk platforms within operational enterprise contexts. The implemented systems demonstrate capacity support for millions of geographic locations, thousands of institutional portfolios, and high-frequency analytics to serve globally distributed organizations. Measurable improvements include performance optimization, scalability enhancement, reliability assurance, and stakeholder adoption metrics, thus validating the successful translation of advanced platform engineering principles into operational impact. The article documents systematic approaches to addressing zero-downtime migration requirements, automated validation protocols, and parallel-run modernization strategies. Results quantify multi-hundred-percent capacity increases, order-of-magnitude scalability improvements, and sub-second response latencies all while ensuring analytical integrity. These achievements provide the frameworks necessary to transform legacy desktop and on-premise risk modeling infrastructure into enterprise-scale cloud platforms for critical business functions across the insurance, banking, and climate risk domains.

The study of quantum information entropy in confined quantum systems plays a pivotal role in quantum information science. In this work, we investigate the quantum information entropy of a particle in a quantum well under an external electric field by calculating the Rényi entropy (), Tsallis entropy () and the Onicescu information energy (). Our findings reveal that these entropic measures exhibit a strong dependence on the entropy order ( α ). Although both and demonstrate a decreasing trend with increasing α for any given quantum state, they differ significantly in their sensitivity to α variations, displaying distinct behaviors that provide unique insights into quantum uncertainty. For the Onicescu information energy (), a sharp transition around α = 1 appears. This characteristic could be significant for designing and analyzing quantum systems, particularly in enhancing their information processing capabilities. By further examining how these entropic measures vary with quantum well width and electric field strength, we gain a deeper understanding of the system's controllability. This empowers researchers to select optimal metrics for specific applications and improve the accuracy of quantum analyses. This study underscores the importance of entropy‐based analysis in quantum mechanics for advancing quantum control, information theory, and technology development.

Measurement and Analysis of Relevant Room Acoustic Parameters in Speech and Language Therapy Rooms and Evaluation of Their Influence on Objective Voice Quality Metrics.

Electrochemical CO 2 reduction to formate using Sn and Bi catalysts was stepwise scaled up and evaluated, enabling the identification of optimal configurations and performance metrics for potential industrial deployment.

BackgroundSodium [18F]Fluoride Positron Emission Tomography (Na[18F]F PET) is a promising imaging biomarker for evaluating bone metabolism in spondyloarthritis (SpA) and other bone affecting diseases. Accurate quantification of tracer uptake is essential for assessing disease activity and treatment response. This study aimed to determine optimal simplified metrics for Na[18F]F uptake and evaluate their performance compared to net influx rate (Ki).ResultsA prospective study included 54 SpA patients undergoing Na[18F]F PET/CT scans at baseline and after 12 weeks of therapy. Dynamic PET in combination with venous blood sampling was analyzed to derive kinetic parameters, including Ki in 43 scans that included pathological uptake in the dynamic field of view. Semi-quantitative standardized uptake values (SUVs) corrected body weight (BW), lean body mass (LBM), body surface area (BSA) and skeletal volume (SV) were compared and correlations between Ki and SUVs were assessed cross-sectionally and longitudinally. Based on analysis of the blood sample data, there was a significant difference between SUV corrected for BW between patients who weighted more and less than 85 kg (p < 0.01 at all sample moments). When LBM or SV was used, this difference disappeared (p > 0.05). There was a significant correlation between Ki and various SUV-metrics, with SUVpeak-LBM at 25–30 min yielding the highest correlation both cross-sectionally (R2 = 0.77, p < 0.01), and longitudinally (R2 = 0.54, p < 0.01).ConclusionsNa[18F]F uptake quantification of lesions in the axial skeleton of SpA patients can be performed cross-sectionally and longitudinally with simplified uptake measures, particularly SUVpeak, normalized using LBM or SV. This offers a more reliable approach to evaluating disease activity and treatment responses compared to BW.Supplementary InformationThe online version contains supplementary material available at 10.1186/s13550-025-01337-0.

PurposeCongenital hypothyroidism (CH) is a common cause of severe intellectual disability, affecting approximately 1 in 2,000 newborns globally. Treatable with early intervention, congenital hypothyroidism has long been a target of newborn screening programs. Current thyroid stimulating hormone (TSH) based programs suffer from low positive predictive value, resulting in unnecessary diagnostic investigations. Congenital hypothyroidism screening has proven challenging for machine learning previously due to massive class imbalance and having a single well known predictor, preventing acceptable screening sensitivity. This study represents the most comprehensive evaluation of machine learning for congenital hypothyroidism screening to date.MethodsAnalyzing data from 616,910 infants screened by Newborn Screening Ontario between 2019 and 2024. 12 classification and 12 resampling algorithms were trained using 4 different optimization metrics, for a total of 576 distinct models evaluated using stratified 5-fold cross-validation to ensure robustness. Models were optimized for sensitivity and then positive predictive value using various metrics. Model explainability was assessed using SHAP values and feature importances.ResultsWe were able to create a model achieving 16.8% PPV while maintaining 100% sensitivity using a RUSBoost classifier and Gaussian Noise resampling. This represents a 60% improvement in positive predictive value over the current approach. TSH remained the dominant predictor as in current screening, but our model was able to include minor amounts of additional information from other features to improve performance.ConclusionThese machine learning algorithms show no missed cases of CH and are able to significantly improve performance across robust testing. The findings suggest that machine learning offers a promising avenue for refining TSH-based CH screening processes, reducing false positives, and alleviating unnecessary stress and costs associated with current methods used by the majority of newborn screening programs globally.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12911-025-03312-0.

High carbon emissions have become a major problem perplexing human society, and the use of shale for CO2 storage has a broad application prospect and significance. In this paper, a SiO2- Al2O3 heterostructure model representing clay shale is established, and the adsorption mechanisms of CO2 in clay shale are studied systematically by means of grand canonical Monte Carlo (GCMC), molecular dynamics (MD) and density functional theory (DFT). The findings indicate that the system's equilibrium shifts toward adsorption as pressure increases, enhancing gas uptake. Conversely, rising temperature favors the desorption equilibrium, thereby reducing the overall adsorption capacity. Higher water content reduces CO2 adsorption capacity in hydrous SiO2-Al2O3 nanopores. For every 5 wt% increment in water content, the CO2 adsorption amount decreases by approximately 18.6%. The density profiles show that the interaction between H2O and adsorption sites on the shale surface is stronger than that of CO2. The radial distribution functions indicate the difference of CO2 distribution between SiO2 region and Al2O3 region and reveal the effect of water on Al2O3 region is greater than that of SiO2 region. This study has an in-depth exploration of the adsorption rule and migration mechanism of CO2 in clay shale, which may contribute preliminary theoretical insights for optimizing CO2 adsorption and storage. The simulation software employed in this study is Materials Studio, and the associated force field utilized is COMPASS III. The adsorption configurations are obtained from the Sorption module and molecular dynamics simulations are performed on it by using the Forcite module with the NVT ensemble. Based on the DFT, molecular optimization and performance metrics analysis are all calculated using the DMol3 module.

ABSTRACT We study how to unwind stochastic order flow with minimal transaction costs. Stochastic order flow arises, e.g., in the central risk book (CRB), a centralized trading desk that aggregates order flows within a financial institution. The desk can warehouse in‐flow orders, ideally netting them against subsequent opposite orders (internalization), or route them to the market (externalization) and incur costs related to price impact and bid‐ask spread. We model and solve this problem for a general class of in‐flow processes, enabling us to study in detail how in‐flow characteristics affect optimal strategy and core trading metrics. Our model allows for an analytic solution in semi‐closed form and is readily implementable numerically. Compared with a standard execution problem where the order size is known upfront, the unwind strategy exhibits an additive adjustment for projected future in‐flows. Its sign depends on the autocorrelation of orders; only truth‐telling (martingale) flow is unwound myopically. In addition to analytic results, we present extensive simulations for different use cases and regimes, and introduce new metrics of practical interest.

Questionnaires are among the oldest and most widely used instruments in practice to measure variables relevant to traits of interest that cannot be easily measured by physical devices, for example, depression. In many clinical settings, the scope of an existing questionnaire is often unfit to apply to a new study population, whose underlying characteristics are different from those of the original population used for the questionnaire's development and/or validation. Motivated by a cohort study of elderly asthma patients, we aim to examine associations between clinical outcomes and quality of life (QoL) measured by a QoL questionnaire. To increase comparability, we consider a supervised learning method to identify a subset of questions whose summary score is strongly associated with a specific clinical outcome under investigation. The resultant set of selected items gives an optimal summary metric of the questionnaire, which improves both statistical power and clinical interpretation. Our item extraction procedure is built upon the best subset algorithm implemented by a mixed integer programming, which enjoys both theoretical guarantee of selection consistency and flexibility of handling nonresponse missing data. Moreover, estimation uncertainty is analyzed by the means of noise perturbation. Our methodology is first evaluated by extensive simulation studies with comparisons to existing methods and then applied to derive tailored QoL scores adaptive to two clinical outcomes of lung function measure (FEV1) and asthma control test (ACT), respectively, among elderly people with persistent asthma.

This research introduces a meticulously designed dual-port antenna tailored to operate within the 2.3-2.4 GHz frequency spectrum. The design specifically addresses challenges of achieving high isolation, reducing envelope correlation, and maintaining robust diversity performance in compact multi-antenna systems. The configuration prioritizes optimal performance metrics to meet stringent communication demands. Notably, the isolation between antenna elements surpasses 20 dB, ensuring minimal interference and enhanced signal integrity in diverse communication environments. Achieving impressive radiation performance, the antenna promises robust signal transmission capabilities while adhering to strict power consumption constraints. Its exceptionally low ECC of less than 0.1 contributes to heightened data reliability, crucial in modern communication systems. Moreover, the antenna exhibits a remarkable diversity gain, nearing 10 dB, facilitating improved signal reception and effectively combating fading and multipath propagation. Notably, its measured channel capacity of 8 bps/Hz underscores its potential for high-bandwidth applications. Fabrication and measurement outcomes meticulously align with theoretical projections, confirming successful synchronization between the antenna's designed specifications and real-world performance, validating its practical viability for diverse wireless communication systems.