Search Results (8,415)

Background/Objectives: Accurate prognostication for patients with chronic critical illness (CCI) following brain injury remains challenging. Conventional scoring systems like the Acute Physiology and Chronic Health Evaluation II (APACHE II) and the Nutrition Risk in the Critically Ill (NUTRIC) score are validated as “left-side” models for risk stratification at intensive care unit (ICU) admission but may not capture the evolving trajectory of prolonged illness. This study aimed to evaluate the prognostic performance of APACHE II and NUTRIC as “right-side” models—assessed at intervals closer to the outcome—by testing the hypothesis that their predictive accuracy for in-hospital mortality improves when measured nearer to the time of death. Methods: In this real-world data analysis study, data were extracted from the electronic health records (Russian Intensive Care Dataset [RICD] v. 2.0) of 328 adult patients with CCI following brain injury. The discriminative ability of repeatedly assessed APACHE II and NUTRIC scores for predicting mortality was analyzed by calculating the area under the receiver operating characteristic curve (AUROC) for three predefined intervals before death: within ≤7 days, 8–14 days, and ≥15 days. Results: Among the 328 patients (median age 64 years; 18.3% in-hospital mortality), a total of 380 paired score assessments were analyzed. The predictive performance for both scores was highest within 7 days of death (APACHE II AUROC: 0.883; NUTRIC AUROC: 0.839). Discriminatory ability declined at 8–14 days (APACHE II AUROC: 0.807; NUTRIC AUROC: 0.778) and was poorest at ≥15 days before death (APACHE II AUROC: 0.671; NUTRIC AUROC: 0.681). The NUTRIC score consistently demonstrated higher AUROC values than APACHE II across all intervals, though the differences were not statistically significant. Conclusions: In patients with CCI following brain injury, the prognostic accuracy of APACHE II and NUTRIC scores is time-dependent, peaking immediately before death and offering poor long-term prediction from admission. These findings underscore the limitation of static, admission-based models and highlight the necessity for developing dynamic, personalized and time-sensitive prognostic tools tailored to the evolving course of chronic critical illness. Full article

Accurate multi-leaf collimator (MLC) motion prediction is a prerequisite for precise dose delivery in advanced techniques such as IMRT and VMAT. Traditional patient-specific quality assurance (QA) methods remain resource-intensive and prone to physical measurement uncertainties. This study aimed to develop machine learning (ML) models to predict delivered MLC positions using kinematic parameters extracted from DICOM-RT plans for the Elekta Versa HD system. A dataset comprising 200 patient plans was constructed by pairing planned MLC positions, velocities, and accelerations with corresponding delivered values parsed from unstructured trajectory logs. Four regression models, including linear regression (LR), were trained to evaluate the deterministic nature of the Elekta servo-mechanism. LR demonstrated superior prediction accuracy, achieving the lowest mean absolute error (MAE) of 0.145 mm, empirically confirming the fundamentally linear relationship between planned and delivered trajectories. Subsequent dosimetric validation using ArcCHECK measurements on 17 clinical plans revealed that LR-corrected plans achieved statistically significant improvements in gamma passing rates, with a mean increase of 2.24% under the stringent 1%/1 mm criterion (p < 0.001). These results indicate that the LR model successfully captures systematic mechanical signatures, such as inertial effects. This study demonstrates that a computationally efficient LR model can accurately predict Elekta MLC performance, providing a robust foundation for implementing ML-based virtual QA. This approach is particularly valuable for time-sensitive workflows like adaptive radiotherapy (ART), as it significantly reduces reliance on physical QA resources. Full article

Energy-efficient Train Control (EETC) strategy needs to meet safety, punctuality, and energy-saving requirements during train operation, and puts forward higher requirements for online use and adaptive ability. In order to meet the above requirements and reduce the dependence on an accurate mathematical model of train operation, this paper proposes a train-speed trajectory-optimization method combining data-driven energy consumption estimation and deep reinforcement learning. First of all, using real subway operation data, the key unit basic resistance coefficient in train operation is analyzed by regression. Then, based on the identified model, the energy consumption experiment data of train operation is generated, into which Gaussian noise is introduced to simulate real-world sensor measurement errors and environmental uncertainties. The energy consumption estimation model based on a Backpropagation (BP) neural network is constructed and trained. Finally, the energy consumption estimation model serves as a component within the Deep Deterministic Policy Gradient (DDPG) algorithm environment, and the action adjustment mechanism and reward are designed by integrating the expert experience to complete the optimization training of the strategy network. Experimental results demonstrate that the proposed method reduces energy consumption by approximately 4.4% compared to actual manual operation data. Furthermore, it achieves a solution deviation of less than 0.3% compared to the theoretical optimal baseline (Dynamic Programming), proving its ability to approximate global optimality. In addition, the proposed algorithm can adapt to the changes in train mass, initial set running time, and halfway running time while ensuring convergence performance and trajectory energy saving during online use. Full article

The diesel substitution ratio is a key parameter influencing the combustion characteristics and energy conversion efficiency of hydrogen diesel dual-fuel free-piston engines. This study develops a thermodynamic hydrodynamic coupled model for a dual-fuel free engine to investigate the effects of five substitution ratios (15%, 20%, 25%, 30%, and 35%) on in-cylinder mixture formation, combustion characteristics, and emission performance. The key novelty of this work lies in employing this fully coupled combustion-dynamics model to systematically optimize the hydrogen–diesel substitution ratio, which explicitly captures the critical feedback between combustion and the piston’s unique motion. The cumulative heat release served as the key quantitative metric. The analyzed parameters included the gas mixture fraction, turbulent kinetic energy, flow trajectories, in-cylinder pressure and temperature, combustion reaction rate, unburned equivalent ratio, cumulative heat release and its rate, heat release rate, and emission mass. The results demonstrate that the engine’s overall performance is optimal at a substitution ratio of 25%. At this ratio, a peak volumetric mixture fraction of 0.0088 was achieved with a broad distribution range, indicating significantly improved spatial fuel uniformity. The flow field exhibited organized swirl patterns that enhanced fuel dispersion. The peak in-cylinder pressure reached 7.2 MPa, which was 0.044 MPa higher than that of the 20% group. The combustion temperature remained stable, with a peak value of 1606 K, exceeding the 20% and 30% groups by 7 K and 16 K, respectively. The heat release phase was well-synchronized with the piston motion, ensuring a high proportion of premixed combustion for thorough fuel oxidation. Although nitrogen oxide (NOx) emissions were slightly higher, the reduction in soot was substantially greater than in the 20% group, leading to overall superior performance compared to the other substitution ratios. This study develops a thermodynamic hydrodynamic coupled model for a dual-fuel free-piston engine by leveraging the interaction between piston motion and combustion. This paper presents a novel strategy for optimizing the substitution ratio in a free piston engine via a fully coupled combustion-dynamics model. Full article

In indoor environments, fusion localization methods that combine Wi-Fi fingerprinting and Pedestrian Dead Reckoning (PDR) are constrained by the high sensitivity of traditional filters, such as the Extended Kalman Filter (EKF), to initial states and by their susceptibility to nonlinear drift. This study presents a Wi-Fi/PDR fusion localization approach based on global geometric alignment optimized via a Genetic Algorithm (GA). The proposed method models the PDR trajectory as an integrated geometric entity and performs a global search for the optimal two-dimensional similarity transformation that aligns it with discrete Wi-Fi observations, thereby eliminating dependence on precise initial conditions and mitigating multipath noise. Experiments conducted in a real office environment (14 × 9 m, eight dual-band APs) with a double-L trajectory demonstrate that the proposed GA fusion achieves the lowest mean error of 0.878 m (compared to 2.890 m, 1.277 m, and 1.193 m for Wi-Fi, PDR, and EKF fusion, respectively) and an RMSE of 0.978 m. It also attains the best trajectory fidelity (DTW = 0.390 m, improving by 71.0%, 14.7%, and 27.8%) and the smallest maximum deviation (Hausdorff = 1.904 m, 52.4% lower than Wi-Fi). The cumulative error distribution shows that 90% of GA fusion errors are within 1.5 m, outperforming EKF and PDR. Additional experiments that compare the proposed GA optimizer with Levenberg–Marquardt (LM), particle swarm optimization (PSO), and Procrustes alignment, as well as tests with 30% artificial Wi-Fi outliers, further confirm the robustness of the Huber-based cost and the effectiveness of the global optimization framework. These results indicate that the proposed GA-based fusion method achieves high robustness and accuracy in the tested office-scale scenario and demonstrate its potential as a practical multi-sensor fusion approach for indoor localization. Full article

Detecting aircraft in Airport Surface Movement Radar (SMR) imagery presents a unique challenge rooted in the conflict between object symmetry and data asymmetry. While aircraft possess strong structural symmetry, their radar signatures are often sparse, incomplete, and highly asymmetric, leading to target loss and position jitter in traditional detection algorithms. To overcome this, we introduce SWCR-YOLO, a keypoint detection framework designed to learn and enforce the target’s implicit structural symmetry from its imperfect radar representation. Our model reconstructs a stable aircraft pose by localizing four keypoints (nose, tail, wingtips) that define its symmetric axes. Based on YOLOv11n, SWCR-YOLO incorporates a MultiScaleStem module and wavelet transforms to effectively extract features from the sparse, asymmetric scatter points, while a Multi-Scale Convolutional Attention (MSCA) module refines salient information. Crucially, training is guided by a Geometric Regularized Keypoint Loss (GRKLoss), which introduces a symmetry-based prior by imposing angular constraints on the keypoints to ensure physically plausible pose estimations. Our symmetry-aware approach, on a real-world SMR dataset, achieves an mAP50 of 88.2% and reduces the trajectory root mean square error by 51.8% compared to MTD-CFAR pipeline methods, from 8.235 m to 3.968 m, demonstrating its effectiveness in handling asymmetric data for robust object tracking. Full article

The Maximum Entropy (MaxEnt) model, integrated with ArcGIS (a geographic information system), was employed to project potential species distribution under current conditions and future climate scenarios (SSP1–2.6, SSP2–4.5, SSP5–8.5) for the 2050s, 2070s, and 2090s. Model optimization involved testing 1160 parameter combinations. The optimized model (FC = LQ, RM = 0.1) exhibited significantly improved predictive performance, with an average AUC of 0.967. Under current conditions, the estimated core suitable habitat spans 35.62 × 10⁴ km², primarily located in southern China. Future projections indicated a non-linear trajectory: an initial contraction of total suitable area by mid-century, followed by a substantial expansion by the 2090s, particularly under high-emission scenarios. Simultaneously, the distribution centroid shifted northwestward. The primary factors influencing distribution were the annual mean temperature (Bio1, 41.1%) and the precipitation of the coldest quarter (Bio19, 20.0%). These findings establish a critical scientific basis for developing climate-adaptive conservation strategies, including the identification of priority climate refugia in Fujian province, China, and planning for assisted migration to northwestern regions. Full article

Background: Wolfram syndrome (WS) is an ultrarare neuroendocrine disorder caused by pathogenic variants in WFS1, frequently leading to progressive neurological, autonomic, and cognitive impairment. Anticipating neurological trajectories remains challenging due to marked phenotypic variability and limited genotype–phenotype data. Methods: Forty-five genetically confirmed patients with WS were evaluated between 1998 and 2024 in Spain. All WFS1 variants were systematically classified by exon, zygosity, protein-level functional impact, and predicted wolframin production (Classes 0–3). Machine learning models (Random Forests with engineered gene–gene interaction terms) were applied to predict neurological manifestations and identify the strongest genetic determinants of symptom severity. Results: Neurological involvement was present in 93% of patients. The most prevalent manifestations were absence of gag reflex (67%), gait instability (64%), dysphagia (60%), and sialorrhea (60%), followed by dysmetria (56%), impaired tandem gait (53%), anosmia (44%), dysarthria (44%), and adiadochokinesia (42%). Most symptoms emerged in early adulthood (23–26 years), whereas cognitive decline occurred later (29.9 ± 12.2 years). Homozygosity for truncating variants—particularly c.409_424dup16 (Val142fsX110)—and complete loss of wolframin production (Class 0; 67–83% across symptoms) were the strongest predictors of early and severe neurological involvement. Machine learning models achieved high discrimination for ataxia, gait instability, and absent gag reflex (AUC 0.63–0.86; calibrated AUC up to 0.97), identifying Mut1_Protein_Class and Mut2_Protein_Class as dominant predictors across all phenotypes, followed by coherent secondary effects from zygosity × exon interaction terms (Prod_mgm). Conclusions: Integrating detailed genetic classification with machine learning methods enables accurate prediction of neurological outcomes in WS. Protein-level dysfunction and allele interaction structure are the principal drivers of neurological vulnerability. This framework enhances precision diagnosis and offers a foundation for individualized surveillance, clinical risk stratification, and future therapeutic trial design in WFS1-related disorders. Full article

Predicting the Remaining Useful Life (RUL) of engine oil is critical for proactive maintenance and fleet reliability. However, irregular and noisy single-point sampling presents challenges for conventional prognostic models. To address this, a hierarchical physics-informed transfer learning (TL) framework is proposed that reconstructs nonlinear degradation trajectories directly from non-time-series data. The method uniquely integrates Arrhenius-type oxidation kinetics and thermochemical laws within a multi-level TL architecture, coupling fleet-level generalization with engine-specific adaptation. Unlike conventional approaches, this framework embeds physical priors directly into the transfer process, ensuring thermodynamically consistent predictions across different equipment. An integrated uncertainty quantification module provides calibrated confidence intervals for RUL estimation. Validation was conducted on 1760 oil samples from dump trucks, dozers, shovels, and wheel loaders operating under real mining conditions. The framework achieved an average R² of 0.979 and RMSE of 10.185. This represents a 69% reduction in prediction error and a 75% narrowing of confidence intervals for RUL estimates compared to baseline models. TL outperformed the asset-specific model, reducing RMSE by up to 3 times across all equipment. Overall, this work introduces a new direction for physics-informed transfer learning, enabling accurate and uncertainty-aware RUL prediction from uncontrolled industrial data and bridging the gap between idealized degradation studies and real-world maintenance practices. Full article

Terahertz Synthetic Aperture Radar (THz-SAR) is highly sensitive to platform vibrations and trajectory deviations, which introduce severe phase errors and limited resolution. Typically, platform vibrations and trajectory deviations are investigated individually, and vibrations are modeled as a stationary sine term. In this work, a hybrid motion compensation (MOCO) scheme is proposed to address both platform vibrations and trajectory deviations simultaneously, achieving improved imaging quality. The scheme initiates with a parameter self-adaptive quadratic Kalman filter designed to resolve severe phase wrapping. Then, platform vibration is modeled as a non-stationary multi-sine term, whose components are accurately extracted using an improved signal decomposition algorithm enhanced by a dynamic noise adjustment mechanism. Subsequently, the trajectory deviation is parameterized following subaperture division, estimated using a hybrid optimizer that combines particle swarm optimization and gradient descent. Additionally, a composite modulated waveform application ensures low sidelobes and a low probability of intercept (LPI). Extensive simulations on point targets and complex scenes under various signal-to-noise-ratio (SNR) conditions are applied for SAR image reconstruction, demonstrating robust suppression of motion errors. Under identical simulated error conditions, the proposed method achieves an azimuth resolution of 4.28 cm, which demonstrates superior performance compared to the reported MOCO techniques. Full article

Ensuring the stability and accuracy of the output voltage in DC/DC buck converters (DBCs) is critical for reliable operation. This paper investigates an observer-based adaptive command-filtered controller designed for DBC systems subject to lumped disturbances. First, a mathematical model of the system is developed on the basis of switching modes. Then, a simplified extended state observer (SESO) is elaborated to mitigate the effects of lumped disturbances. A command filter technique with an integrated adaptive law is subsequently synthesized to enhance output voltage regulation. The stability of the observer and DBC control system is rigorously certified using the Lyapunov principle. Finally, simulation and experimental approaches are exploited to confirm the validity of the proposed method. Compared to state-of-the-art approaches, the proposed observer-based adaptive command-filtered controller improves tracking performance by 96.1% and 77.8% in simulations and 84.4% and 49.1% in experiments under a sinusoidal reference trajectory. Full article

This study proposes an automatic denatured recognition method of biological tissue during high-intensity focused ultrasound (HIFU) therapy. The technique integrates ultrasonic phase space reconstruction (PSR) with a convolutional block attention mechanism-enhanced EfficientNet-B0 model (CBAM-EfficientNet-B0). Ultrasonic echo signals are first transformed into high-dimensional phase space reconstruction trajectory diagrams using PSR, which reveal distinct fractal and chaotic characteristics to analyze tissue complexity. The CBAM module is incorporated into EfficientNet-B0 to enhance feature extraction from these nonlinear dynamic representations by focusing on critical channels and spatial regions. The network is further optimized with Dropout and Scaled Exponential Linear Units (SeLUs) to prevent overfitting, alongside a cosine annealing learning rate scheduler. Experimental results demonstrate the superior performance of the proposed CBAM-EfficientNet-B0 model, achieving a high recognition accuracy of 99.57% and outperforming five benchmark CNN models (EfficientNet-B0, ResNet101, DenseNet201, ResNet18, and VGG16). The method avoids the subjectivity and uncertainty inherent in traditional manual feature extraction, enabling effective identification of HIFU-induced tissue denaturation. This work confirms the significant potential of combining nonlinear dynamics, fractal analysis, and deep learning for accurate, real-time monitoring in HIFU therapy. Full article

Point-of-Interest (POI) recommendation predicts users’ future check-ins based on their historical trajectories and plays a key role in location-based services (LBS). Traditional approaches such as collaborative filtering and matrix factorization model user–POI interaction matrices fail to fully leverage spatio-temporal information and semantic attributes, leading to weak performance on sparse and long-tail POIs. Recently, Graph Neural Networks (GNNs) have been applied by constructing heterogeneous user–POI graphs to capture high-order relations. However, they still struggle to effectively integrate spatio-temporal and semantic information and enhance the discriminative power of learned representations. To overcome these issues, we propose Spatio-Temporal and Semantic Dual-Channel Contrastive Alignment for POI Recommendation (S²DCRec), a novel framework integrating spatio-temporal and semantic information. It employs hierarchical relational encoding to capture fine-grained behavioral patterns and high-level semantic dependencies. The model jointly captures user–POI interactions, temporal dynamics, and semantic correlations in a unified framework. Furthermore, our alignment strategy ensures micro-level collaborative and spatio-temporal consistency and macro-level semantic coherence, enabling fine-grained embedding fusion and interpretable contrastive learning. Experiments on real-world datasets, Foursquare NYC, and Yelp, show that S²DCRec outperforms all baselines, improving F1 scores by 4.04% and 3.01%, respectively. These results demonstrate the effectiveness of the dual-channel design in capturing both sequential and semantic dependencies for accurate POI recommendation. Full article

Background/Objective: Hip fractures in older adults are a major public health concern due to their high rates of morbidity, mortality, and long-term disability. Although surgical and postoperative care have improved, recovery outcomes remain highly variable. Identifying early functional recovery patterns could support individualized rehabilitation and discharge planning. This study aimed to identify distinct early recovery trajectories of motor function within 30 days after hip fracture surgery using growth mixture modeling (GMM) and to examine patient- and hospital-level factors associated with these patterns. Methods: A retrospective cohort study was conducted using data from the Nagano Hip Fracture Database, including 2423 patients aged ≥65 years across 17 acute care hospitals in Japan (2019–2024). Functional recovery was measured using the motor subscale of the Functional Independence Measure (FIM-motor) at 0, 7, and 28 days post-admission. Latent trajectory model was used to identify distinct recovery patterns. Multinomial logistic regression analyzed predictors of class membership. Results: Three recovery trajectories were identified: high/rapid improvement (26.7%), intermediate (32.6%), and poor/flat recovery (40.7%). Older age, cognitive impairment, and lower baseline mobility were strongly associated with membership in the poor-recovery class. Early trajectory classes significantly predicted discharge outcomes, including FIM-motor scores and discharge destination. Sensitivity analysis confirmed the robustness of findings, with minimal impact from hospital-level clustering. Conclusions: Distinct early recovery trajectories exist after hip fracture surgery and are strongly influenced by baseline cognitive and functional status. Early identification of recovery patterns can enhance personalized rehabilitation and inform discharge planning, offering valuable insights for clinical practice. Full article

This paper develops a mathematical approach to the analysis of the stability of economic equilibria in nonsmooth models. The $\lambda$-Hölder apparatus of subdifferentials is used, which extends the class of systems under study beyond traditional smooth optimization and linear approximations. Stability conditions are obtained for solutions to intertemporal choice problems and capital accumulation models in the presence of nonsmooth dependencies, threshold effects, and discontinuities in elasticities. For $\lambda$-Hölder production and utility functions, estimates of the sensitivity of equilibria to parameters are obtained, and indicators of the convergence rate of trajectories to the stationary state are derived for $\lambda >1$. The methodology is tested on a multisectoral model of economic growth with technological shocks and stochastic disturbances in capital dynamics. Numerical experiments confirm the theoretical results: a power-law dependence of equilibrium sensitivity on the magnitude of parametric disturbances is revealed, as well as consistency between the analytical $\lambda$-Hölder convergence rate and the results of numerical integration. Stochastic disturbances of small variance do not violate stability. The results obtained provide a rigorous mathematical foundation for the analysis of complex economic systems with nonsmooth structures, which are increasingly used in macroeconomics, decision theory, and regulation models. Full article