Search Results (4,237)

Rotating machinery prognostics is essential for ensuring the reliability and operational safety of industrial systems. Although gated recurrent units (GRUs) have achieved competitive performance in remaining useful life (RUL) prediction, whether their internal dynamics are consistent with irreversible degradation mechanisms remains largely unexplored. This study revisits GRU-based prognostics from a physics-consistency perspective and analyzes the potential mismatch between gating mechanisms and degradation evolution. A full-life benchmarking framework is developed based on the XJTU-SY bearing run-to-failure dataset. A training-based health indicator (HI) is constructed through multi-domain vibration feature extraction and principal component analysis, where the degradation-state representation and RUL prediction objective are explicitly distinguished to avoid physically inconsistent supervision. Several representative approaches, including statistical models and deep learning architectures (LSTM, GRU, TCN, and Transformer), are evaluated using both prediction accuracy metrics (RMSE, MAE, and R²) and physical consistency criteria (monotonicity index, monotonicity violation index, and degradation trend consistency). Experimental results demonstrate that superior prediction accuracy does not necessarily guarantee physically consistent degradation modeling. Although GRU provides competitive RUL prediction performance, its hidden-state evolution and gating responses exhibit noticeable non-monotonic behaviors during degradation progression. These findings reveal a potential discrepancy between prediction-oriented recurrent learning mechanisms and irreversible degradation dynamics, highlighting the importance of incorporating physics-consistency evaluation into reliable data-driven prognostic models. Full article

Federated learning (FL) enables collaborative model training across decentralized clients while preserving data privacy, yet real-world deployments still face communication bottlenecks, performance degradation under heterogeneous client data, and limited personalization. In this study, we introduce FedMCP++, a modular and communication-efficient personalized FL framework in which every client owns a complete private model—a lightweight convolutional backbone with a private expert head—and collaboration is carried out entirely through knowledge exchange rather than parameter exchange. In each round, clients share only temperature-softened class predictions and class-wise feature prototypes computed on a small public proxy set; the server fuses them into an accuracy-weighted teacher and broadcasts the result, and clients realign their models through knowledge distillation, an instance-level contrastive objective, and prototype alignment. We evaluate FedMCP++, its ablations, and two knowledge-based baselines on six benchmark vision datasets with 10, 20, and 30 clients. The results indicate dataset-dependent trade-offs rather than uniform superiority: collaborative distillation improves average client-level accuracy over independent local training in twelve of eighteen configurations—most clearly under severe per-client data scarcity (e.g., up to +2.7 percentage points on KMNIST and +2.4 on STL-10 with 20–30 clients)—whereas independent training ensembles remain strongest on SVHN and CIFAR-10 at the studied budgets. Because no parameters are transmitted, the per-round uplink payload is a fixed-size 42.6 KB message, 9.9–12.8× smaller than full-model synchronization, and is invariant to model capacity. These properties make FedMCP++ a flexible framework for personalized FL in wireless edge and Internet of Things environments where bandwidth and privacy constraints are paramount. Full article

Although asymbiotic orchid seed germination was first achieved in vitro in 1922, the prediction of germination requirements under in vitro conditions still remains complicated. To address this, we developed a machine learning framework to classify the ex situ asymbiotic germination potential of wild orchids into four discrete groups: Low (0–30%), Mid (31–50%), High (51–80%), and Max (81–100%). Models were trained on a dataset of 203 species, utilizing seed morphometrics—specifically, the embryo-to-testa (E:S) length ratio—alongside core ecological traits (subfamily, growth habit, habitat, and climate zone), as well as chemical scarification duration as a proxy of seed permeability. Validation leveraged novel germination and trait data from 26 taxa from Greece (17) and Armenia (9), published here for the first time. To mitigate class imbalance and prevent algorithmic bias toward highly germinating species, we applied inverse frequency weighting during training. Iterative testing of six algorithms revealed that the “Step 4” feature matrix (excluding climate zone and pretreatment duration) yielded the optimal predictive balance. K-Nearest Neighbor (KNN) and Support Vector Machine (SVM) emerged as the superior models, achieving overall accuracies of 44.4% and 61.1%, respectively, with both achieving 100% accuracy for low-germinating species. Finally, we synthesized a novel database compiling new seed morphometrics from Armenia (17 taxa), Greece (52 taxa), and the data from the literature (479 taxa). After filtering previously utilized species, we generated a prediction pool of 361 orchid taxa. Applying our Step 5 KNN and SVM models to forecast their germination behavior revealed distinct variations linked to ecological profiles. This high-accuracy framework, particularly for low-germinability groups, offers a powerful screening tool for ex situ conservation planning. The final trained models are compiled in the publicly available R (v. 4.6.0) package OrchidGermClass. Full article

This research aims to investigate and improve the accuracy of porosity prediction in steel ingot casting by leveraging Thercast finite element simulations. In particular, the study refines the Niyama criterion through additional physical parameters and solidification modeling, aiming to reduce shrinkage porosity and enhance the overall mechanical reliability of cast components. Simulation results reveal that a lower thermal conductivity and faster cooling rates exacerbate shrinkage porosity, while a refined Niyama indicator using explicit solid-fraction weighting, with viscosity, alloy composition, and shrinkage accounted for through the underlying THERCAST material model, improves spatial localization of porosity-prone regions in the investigated case. For the investigated configuration, reducing the cooling rate to around 1.25 K/s decreased the extent of the simulated region classified as porosity-prone relative to the reference case. Furthermore, the analytical porosity–strength relation indicates a material-dependent reduction in strength when the porosity fraction exceeds 2%, underscoring the structural significance of internal voids. This study extends the practical interpretation of the classical Niyama criterion by combining solid-fraction weighting with material-dependent thermophysical inputs, addressing gaps in existing shrinkage porosity models. The approach integrates simulation findings with actual casting defects identified through ultrasonic scanning and metallographic analysis. By merging experimental insights with advanced finite element simulations, foundries can better regulate casting conditions, particularly cooling rates and thermal gradients, to minimize porosity. The refined porosity prediction framework aids in process optimization, improved material utilization, and superior quality assurance of steel ingots. Full article

Background/Objectives: Preterm birth remains the leading cause of neonatal morbidity and mortality, with cervical insufficiency being a major risk factor. Quantitative fetal fibronectin (qfFN) and the QUiPP App version 2.0 are established tools for preterm birth prediction; however, the value of repeated assessments remains unclear. This study aimed to evaluate the prognostic accuracy of longitudinal qfFN measurements and QUiPP risk assessment in women with progressive cervical insufficiency and to evaluate the prognostic performance of repeated assessments. Methods: In this retrospective cohort study, 88 women with progressive cervical insufficiency who underwent repeated qfFN measurements were included. Cervical length, qfFN, and QUiPP risk were assessed at diagnosis and at subsequent time points. Prognostic performance for delivery within one week and before 34 and 37 gestational weeks was evaluated. Results: The prognostic performance represented by the area under the curve increased from 0.71 to 0.75 for qfFN and from 0.74 to 0.82 for QUiPP from the first to the final assessment, although neither difference reached statistical significance (qfFN p = 0.522; QUiPP risk p = 0.135). For imminent preterm birth, qfFN demonstrated excellent rule-out performance, with a negative predictive value of 100% at <10 ng/mL. QUiPP risk provided superior risk stratification and achieved excellent prognostic accuracy for delivery before 34 weeks at the final assessment (area under the curve = 0.82). Receiver operating characteristic analysis identified optimal cutoffs of approximately 40 ng/mL for qfFN and 23% for QUiPP risk in predicting delivery before 34 weeks. Conclusions: Repeated qfFN assessment combined with QUiPP risk estimation was associated with numerically higher discriminatory performance at the final assessment. While the identified optimal cutoff values need to be validated in prospective trials, they might have the potential to safely reduce unnecessary interventions in the future. qfFN functioned as an effective rule-out test, whereas QUiPP enhanced risk stratification, supporting their combined repeated use to guide clinical management before 34 + 0 weeks of gestation. Full article

Osteonecrosis of the femoral head (ONFH) is a progressive and potentially disabling condition caused by compromised blood supply to the femoral head, leading to bone necrosis and collapse. Early diagnosis is essential to enable joint-preserving interventions and improve patient outcomes. Magnetic resonance imaging (MRI) is currently considered the most sensitive modality for early detection, whereas computed tomography (CT) provides superior assessment of subchondral bone integrity and structural collapse. In recent years, artificial intelligence (AI), particularly machine learning (ML) and deep learning (DL) techniques, has emerged as a promising tool to enhance diagnostic accuracy, automate lesion segmentation, and predict disease progression. This review aims to provide an overview of current AI applications in MRI for ONFH, focusing on early diagnostic, disease staging and classification, volumetric assessment, differential diagnosis and prognostic prediction. A total of 61 articles were initially identified, of which 13 studies (2021–2025) met the inclusion criteria. Results indicate that DL models, particularly convolutional neural networks (CNNs), achieve excellent diagnostic performance, with reported accuracies up to 98.4% and area under the curve (AUC) values reaching 0.98 for early-stage detection. Several models demonstrated performance comparable to or exceeding that of experienced clinicians, particularly in differentiating ONFH from other hip pathologies and in early disease recognition. AI algorithms also showed high accuracy in staging and classification (AUC up to 99.7% in internal validation), as well as in automated segmentation and volumetric assessment (Dice coefficients up to 0.89), enabling objective quantification of necrotic lesions. Furthermore, prognostic models integrating radiomics and ML techniques demonstrated promising results in predicting femoral head collapse (AUC up to 0.85). From a clinical perspective, AI appears to function primarily as a supportive tool, improving diagnostic consistency, efficiency, and reproducibility, and acting as a “second reader” capable of reducing variability among less experienced clinicians. However, significant limitations remain, including dataset heterogeneity, predominance of retrospective and monocentric studies, and limited integration of clinical data. In conclusion, AI-based MRI analysis shows strong potential to enhance the diagnosis, staging, and management of ONFH. Future research should focus on multicenter prospective validation, integration of multimodal clinical data, and development of explainable and generalizable models to facilitate widespread clinical adoption. Full article

Accurate wind speed forecasting plays a crucial role in the safe and stable operation of wind farms and the efficient integration of renewable energy into modern power systems. However, wind speed exhibits complex spatiotemporal variations affected by diverse meteorological conditions, making high-precision prediction a long-standing bottleneck in wind power scheduling. This paper develops a U-shaped spatiotemporal feature fusion network named U-STNet, which realizes joint modeling of inter-turbine spatial correlations and multi-period long-range temporal dependencies. The model maps raw wind speed series to high-dimensional embeddings and adopts an encoder–decoder U-shaped architecture to complete feature encoding, reconstruction and multi-scale feature extraction, which effectively captures the inherent periodic and seasonal patterns of wind speed. Experiments on the SDWPF wind farm dataset show that U-STNet obtains competitive prediction accuracy across all multi-step forecasting horizons. Compared with traditional statistical models, recurrent neural networks and state-of-the-art Transformer baselines, the proposed method exhibits more stable error accumulation characteristics and superior long-step prediction performance. This verifies the effectiveness of jointly modeling turbine spatial topology and multi-scale temporal dependencies for wind speed forecasting. Full article

Gravity-matching navigation is one of the important means for the covert navigation of underwater vehicles. The production and application of gravity reference maps as core key technologies have a very significant impact on the accuracy of underwater navigation. Firstly, this paper adopts the factor analysis method to obtain the comprehensive results of nine characteristic parameters such as the standard deviation and roughness of the gravity reference map by setting a range sliding window. Secondly, the TERCOM algorithm is introduced to conduct simulation verification calculations within the sliding window. After comparing and verifying with the comprehensive results of the factor analysis characteristic parameters, the limitations of statistical methods in the evaluation of the adaptability of gravity reference maps are analyzed. Thirdly, intelligent processing methods such as the learning vector quantization neural network algorithm and the extreme learning machine are proposed. The characteristic parameters of some sliding window gravity reference maps and the simulation verification results of the TERCOM algorithm are used as training samples to predict the adaptability evaluation effect of underwater gravity navigation for other sliding windows. The results show that the prediction results are generally in good agreement with the simulation verification results of the TERCOM algorithm. Compared with the learning vector quantization neural network algorithm, the extreme learning machine algorithm exhibits superior performance in terms of classification accuracy and computational efficiency. Full article

Achieving high fault-classification accuracy alone does not guarantee reliable autonomous operation under varying operating conditions, raising the need to assess prediction reliability and deployment readiness. This work proposes a resilience-oriented framework for fault diagnosis under varying operating conditions, characterizing diagnostic behavior under operating-condition shifts and providing complementary information on confidence, deployability, and supervision requirements. The framework fuses multi-sensor vibration and motor current signals within a Multi-Stage architecture combining a data-driven branch (DD-MSCNN) and a physics-aware branch (PA-MSCNN) integrating order-tracking descriptors, augmented by a confidence-aware deployability assessment layer. Evaluated on the Paderborn KAT dataset across six bidirectional shifts involving speed, torque, and radial force, the results reveal that operating-condition shifts are not equivalent and that their impact is strongly direction-dependent. Physical knowledge does not systematically guarantee superior performance, highlighting the complementary roles of the two representations. To formalize these observations, the Physics Contribution Index (PCI), the Shift Directionality Index (SDI), and a four-level deployability classification are introduced, providing quantitative insights into prediction reliability and autonomous operation readiness in dynamic industrial environments. Full article

Thalassemia represents the world’s most prevalent inherited hemoglobin disorder, affecting approximately 4.4 per 10,000 live births globally. Accurate genetic characterization is indispensable both for definitive diagnosis and for lifetime clinical monitoring. The past two decades have witnessed a paradigm shift from conventional protein-based assays toward comprehensive molecular techniques, including next-generation sequencing (NGS) and third-generation (long-read) sequencing, which in turn have enabled reproductive applications such as preimplantation genetic testing for monogenic disease (PGT-M) to identify unaffected embryos before implantation. (1) To systematically evaluate the molecular techniques available for confirming the diagnosis of alpha- and beta-thalassemia, including their diagnostic accuracy, indications, and limitations; (2) to examine how genotype–phenotype correlation and genetic modifier profiling inform clinical prognosis and therapeutic decision-making; and (3) to define evidence-based genetic monitoring parameters for longitudinal follow-up of patients receiving transfusions, iron chelation, and novel curative therapies including gene therapy. A comprehensive narrative review was conducted by systematically searching PubMed/MEDLINE for English-language peer-reviewed articles published between January 2000 and December 2024. Forty-three studies were ultimately included after applying predefined inclusion and exclusion criteria. Quality of included studies was assessed using SANRA (Scale for the Assessment of Narrative Review Articles). HPLC and capillary electrophoresis remain first-line phenotyping tools; DNA-based confirmation is mandatory for complete genotyping. Among known, previously characterized mutations, NGS-based targeted panels achieve > 95% detection sensitivity, but they require MLPA co-testing or long-read sequencing to detect structural variants such as large deletions. Genotype–phenotype prediction is substantially improved, though not rendered fully deterministic, by profiling three major modifier loci: XmnI (Gγ), BCL11A, and HBS1L-MYB. PGT-M using NGS achieves near-complete genotyping accuracy (>99%) with live birth rates of 40–60% per frozen embryo transfer cycle. For patients receiving curative gene therapy (exagamglogene autotemcel/Casgevy), molecular follow-up protocols spanning 15 years are now recommended. Cardiac T2* MRI remains the most reliable non-invasive tool for iron overload follow-up, superior to serum ferritin alone. A tiered, genotype-informed approach—combining HPLC/CE phenotyping, targeted molecular diagnostics, genetic modifier profiling, and periodic re-evaluation—optimizes diagnostic precision and guides individualized management across the thalassemia spectrum. Integration of PGT-M and long-read sequencing into standard care pathways, alongside robust gene therapy follow-up protocols, will define the next era of thalassemia genetics. Full article

Endemic channels are widely used in dengue surveillance to detect unusual increases in case counts. However, conventional approaches that rely on historical averages with fixed standard deviation (SD)-based multipliers may produce unstable thresholds and false alerts. This study compared a conventional SD-based endemic channel with a log-scale SD-based endemic channel incorporating an enhanced alert rule to identify the optimal approach for routine dengue surveillance in Malaysia. Weekly national dengue case data from 2014 to 2024 were analyzed. A rolling validation approach was used to evaluate outbreak detection performance from 2017 to 2023 using three-year historical baselines. Sensitivity, specificity, positive predictive value, negative predictive value, accuracy, and the Youden Index were calculated and pooled across the validation years. The optimal multiplier was selected based on the highest pooled Youden Index using the rolling validation period (2017–2023). The selected approach was then applied independently to the 2024 surveillance data as an operational demonstration of its potential use in routine dengue surveillance. A total of 260 epidemiological weeks were analyzed, of which 42 (16.2%) were classified as outbreak weeks. The log-scale SD-based endemic channel incorporating an enhanced alert rule achieved the highest pooled Youden Index of 0.43 at the optimal multiplier of 0.50. At this multiplier, sensitivity was 0.60, specificity 0.82, positive predictive value 0.35, negative predictive value 0.93, and overall accuracy 0.79. Compared with the conventional SD-based endemic channel at its optimal multiplier (1.00), the proposed approach demonstrated improved sensitivity (0.60 vs. 0.56), specificity (0.82 vs. 0.70), positive predictive value (0.35 vs. 0.23), negative predictive value (0.93 vs. 0.91), accuracy (0.79 vs. 0.68), and Youden Index (0.43 vs. 0.26). The log-scale SD-based endemic channel incorporating an enhanced alert rule demonstrated superior overall outbreak detection performance and was selected as the optimal approach for routine dengue surveillance. Full article

As the core power equipment of the deep-sea mining system, the deep-sea mineral lifting pump continuously operates under harsh service conditions coupled with high hydrostatic pressure, complex marine environments, and solid particle media. Its operating state shows obvious strong nonlinearity, time-varying fluctuation, and local abrupt change features. Therefore, time series health state prediction of the deep-sea mineral lifting pump is of vital engineering significance for realizing predictive maintenance and ensuring the safety of offshore trials and mining operations. Taking the 500 m-level offshore sea trial conducted in the Xisha area of the South China Sea as the engineering background, four critical health characteristic parameters, including shaft power, pump efficiency, motor winding temperature, and outlet radial vibration, are selected to construct a hybrid CNN-LSTM time series prediction model. Comprehensive model evaluation metrics and ablation comparison experiments are adopted to analyze the multi-step-ahead prediction performance of the proposed model. The results show that the CNN-LSTM model achieves optimal comprehensive evaluation indices in one-step prediction and possesses excellent tracking capability for inflection points and amplitude fluctuations of time series data. Although the prediction accuracy decreases gradually with the increase in prediction steps, the model can still effectively characterize the evolutionary trend of pump operating states, and its overall prediction performance is significantly superior to that of single models. This study provides model support and technical reference for the health evaluation, early fault alarm, and maintenance optimization of deep-sea mineral lifting pumps in offshore trials. Full article

With the acceleration of global urbanization and the large-scale development of underground spaces, underground engineering faces extremely complex and variable geological environments and high-risk construction disturbances. The widespread application of the Internet of Things and novel sensing technologies has given rise to structural health monitoring data increasingly characterized by massive volume, high dimensionality, multi-source heterogeneity, and strong spatiotemporal coupling. Traditional data processing methods based on mechanical analysis, empirical formulas, or numerical simulation have increasingly exposed limitations of insufficient accuracy, lengthy computation times, and weak generalization capability when confronted with such engineering big data. Machine learning and deep learning technologies, by virtue of their superior nonlinear mapping capability, advantages in feature extraction from massive data, and flexible architectural design, provide solutions for efficient knowledge extraction and intelligent assessment of underground engineering monitoring data. This paper reviews the current application status and frontier advances of machine learning technologies in the field of underground engineering monitoring data processing in recent years. First, the development trajectory of analytical algorithms evolving from classical shallow machine learning, through temporal and spatial deep learning, to physics-data dual-driven approaches is delineated. Second, targeting the critical challenges of missing field data and sparse sensor deployment, spatiotemporal fusion imputation techniques and spatial reconstruction methods incorporating mechanical prior knowledge are thoroughly evaluated, elucidating the paradigm shift in monitoring philosophy from discrete point-based alarming to inference-augmented sparse sensing that approximates full-field state awareness through model-dependent estimation rather than direct measurement. Third, the applications of machine learning in underground structural deformation mechanism interpretation, key influencing factor identification based on explainable artificial intelligence (AI), and rapid back-analysis of geomechanical parameters are summarized. Finally, composite network architectures and physics-constrained guidance strategies for non-stationary deformation time series prediction under complex and variable working conditions are discussed. A methodological audit of the 73 included studies—of which 33 enter the quantitative comparison tables—reveals that 26 of the 33 audited studies (78.8%) validate exclusively on single-project data, only 1 study conducts rigorous out-of-distribution generalization testing, and none of the 33 studies (0%) provides uncertainty quantification. These findings highlight cross-project generalization and probabilistic prediction as important methodological challenges. This paper aims to provide theoretical references and methodological guidance for safety early warning, intelligent construction, and full life-cycle health management of underground engineering. Full article

Dendrobium polysaccharide is an important indicator for evaluating the quality of Dendrobium huoshanense. To improve the prediction accuracy of polysaccharide content, this study proposes an improved Artificial Lemming Algorithm (IALA) optimized BP neural network model. In IALA, a periodic mutation strategy and a fast hybrid opposition learning strategy (FHOBL) are introduced to enhance population diversity, improve global search ability, and avoid premature convergence. The proposed IALA was first evaluated on CEC2017 and CEC2020 benchmark functions. Experimental results show that IALA achieves better or competitive performance compared with seven other algorithms in terms of mean fitness, best fitness, and standard deviation. Statistical tests, including Wilcoxon rank-sum and Friedman tests, further verify the significant superiority and robustness of IALA. Then, IALA was used to optimize the initial weights and thresholds of BP neural networks for Dendrobium polysaccharide content prediction. The results show that IALA-BP achieves the best overall prediction performance, with an R² of 0.8731, RMSE of 2.1581, and MSE of 4.6683. Compared with standard BP and other optimized BP models, IALA-BP provides more accurate and stable prediction results. Therefore, the proposed IALA-BP model is effective for rapid prediction of Dendrobium polysaccharide content. Full article

Lithology prediction is a fundamental component of geological exploration and hydrocarbon reservoir characterization, playing a critical role in improving subsurface structural interpretation and enhancing resource prediction accuracy. However, well log data are typically characterized by high dimensionality, strong nonlinearity, severe class imbalance, and asymmetric geological feature distributions, which significantly restrict the predictive accuracy and generalization capability of conventional machine learning methods. To address these challenges, this study proposes a symmetry-aware lithology classification framework based on a Hybrid Grey Wolf Optimizer–Genetic Algorithm optimized Extreme Gradient Boosting (GGA-XGB) model. The proposed framework establishes a symmetric collaborative optimization mechanism by integrating the global exploration capability of the Grey Wolf Optimizer (GWO) with the local exploitation ability of the Genetic Algorithm (GA), thereby achieving a balanced optimization strategy between exploration and exploitation. Specifically, GWO first performs coarse-grained global hyperparameter optimization of XGBoost to improve search efficiency and optimization stability, while GA subsequently refines the parameter space to further enhance local optimization accuracy. Experimental results on a multi-class well logging dataset demonstrate that the proposed method achieves outstanding classification performance, with precision, recall, and F1-score all reaching 0.9862. Compared with several conventional machine learning methods, the proposed GGA-XGB framework exhibits superior predictive accuracy. The symmetry-aware optimization strategy provides an effective solution for intelligent lithology prediction under complex geological conditions and offers both theoretical insights into symmetry-aware optimization mechanisms and practical value for intelligent geoscience and hydrocarbon exploration. Full article