Search Results (13,200)

Soil salinisation and periodic flooding jointly shape coastal wetland ecosystems, but how coexisting halophytes differ in their rhizosphere bacterial communities under these conditions remains insufficiently understood. Here, we investigated rhizosphere bacterial communities associated with Suaeda glauca Bunge and Tamarix chinensis Lour. in saline–alkaline habitats of the Yellow River Delta, China. Twenty quadrat-level rhizosphere samples were collected across four plant–habitat groups, and near-full-length 16S rRNA gene amplicons were sequenced using Pacific Biosciences single-molecule real-time sequencing. Our analysis revealed that hydrological habitat and plant identity together contributed to differences in rhizosphere bacterial community composition. Across the dataset, 2325 bacterial operational taxonomic units were identified. T. chinensis showed higher Shannon and Gini–Simpson diversity, whereas richness patterns depended on habitat and the metric examined. Meanwhile, exploratory genus-level association networks revealed host- and habitat-dependent differences in node number, network density and average degree. PICRUSt2-based functional prediction suggested contrasting predicted functional response patterns: the S. glauca rhizosphere showed 19 significantly altered predicted pathways between flooded and non-flooded habitats, whereas the T. chinensis rhizosphere showed no significant pathway shifts after multiple-testing correction. These findings suggest that coexisting halophytes are associated with divergent rhizosphere bacterial community patterns under saline–alkaline and flooding-associated habitat conditions. Full article

A robust fault-tolerant predictive asynchronous switching control method based on robust positively invariant sets is proposed for multi-phase batch processes subject to actuator faults, unknown disturbances, and asynchronous switching. To attenuate the effect of unknown disturbances, a min–max performance index is constructed, by which the robust control problem is formulated as a min–max optimization problem under worst-case disturbance conditions. To improve fault tolerance, robust positively invariant sets are introduced into the controller design so that the system states can remain within a constraint-satisfying feasible region under admissible actuator faults. Moreover, an online pre-switching mechanism is developed to address the phase mismatch between the system phase and controller. By updating the switching timing according to the real-time operating state, the controller can be adjusted to the corresponding control law before the system enters the next phase, thereby reducing the mismatched interval and suppressing state deviation. A case study on the injection and holding phases of the injection molding process shows that the proposed method improves tracking accuracy and operational smoothness under actuator faults, unknown disturbances, and asynchronous switching, demonstrating its effectiveness and applicability. Full article

Background/Objectives: Prehospital identification of non-ST-elevation acute coronary syndrome remains unresolved, and available risk scores incorporate troponin categorically. We compared the discriminative performance of cardiac troponin I measured in a blood sample obtained at first prehospital medical contact, analysed as a continuous variable, against the HEART, GRACE, and TIMI scores for predicting 30-day major adverse cardiac events. Methods: We conducted a prospective cohort study in 333 adults with non-traumatic chest pain suggestive of non-ST-elevation acute coronary syndrome, which was attended by two physician-staffed advanced life support units. Venous blood was drawn at first medical contact and analysed later in the laboratory; the results were not available during patient care. All three scores used this same determination as their troponin component. The primary outcome was a composite of all-cause mortality, type 1 myocardial infarction, and unplanned coronary revascularisation at 30 days. Results: The primary outcome occurred in 82 patients (24.6%). Prehospital troponin discriminated better (area under the curve 0.916, 95% CI 0.875–0.952) than TIMI (0.817), GRACE (0.767), and HEART (0.721); all comparisons were p < 0.001. An exploratory rule-out threshold of 80 ng/L identified 61 patients (18.3%) with no events (sensitivity and negative predictive value 100%, exact 95% CI 95.6–100 and 94.1–100), whereas the ≥780 ng/L stratum showed an event rate of 69.4%. Conclusions: A blood sample obtained at first prehospital medical contact carries substantial prognostic information that clinical scores discard by categorising it. Because it was analysed subsequently in the laboratory, these findings do not evaluate a real-time strategy; they support development and validation of point-of-care devices. Full article

Organoid and organ-on-chip technologies are rapidly evolving platforms for viral research that integrate stem cell biology, tissue engineering, and microfluidics to recapitulate key structural, mechanical, biochemical, and cellular features of human and animal physiology. By incorporating multicellular organoids into perfused microfluidic systems, these models can provide complex, dynamic, and physiologically relevant micro-environments for investigating virus–host interactions that are difficult to capture in conventional two-dimensional cultures and static organoids. Controlled flow, shear stress, extracellular matrix organization, tissue–tissue interfaces, and multicellular signaling enable mechanistic investigation of viral infectivity, dissemination, tissue injury and immune activation. Integration of real-time imaging and biosensors further permits longitudinal monitoring of viral replication, host responses, and tissue integrity, expanding the potential of these platforms for antiviral drug discovery. Recent organoid-on-chip studies using brain, skin, vaginal, respiratory, and intestinal models have demonstrated how tissue architecture, mechanical forces, glycocalyx dynamics, and immune–stromal interactions influence viral tropism and pathogenesis. In this review, we provide a mechanistic and translational overview of organoid and organ-on-chip technologies for studying viral infections, with particular emphasis on models of herpes simplex virus (HSV)-mediated disease. We further examine advances in immune integration, multi-organ systems, biosensing, and computational approaches that are expanding the complexity and predictive potential of these models. Importantly, patient-derived organoids and organ-on-chip platforms can capture interindividual differences in viral susceptibility, host responses, and therapeutic efficacy, providing pharmaceutical research with more precise, patient-relevant data to support drug prioritization and precision antiviral medicine. Finally, we discuss key barriers to broader adoption, including organoid maturation, biological and technical variability, reproducibility, scalability, biosafety, cost, standardization, and regulatory validation. Collectively, these advances position organoid and organ-on-chip technologies as powerful human-relevant models that bridge reductionist in vitro systems and human disease, while continued optimization, standardization, and validation will be essential to realize their full potential for mechanistically informed antiviral discovery, therapeutic development, and precision medicine. Full article

In the testing of aircraft cargo smoke detectors, surrogate smoke is often used in place of fire-generated smoke to avoid the hazards of live-fire experiments. Reproducing the time-varying concentration profile of real fire smoke requires feedback control of the surrogate smoke concentration. Two obstacles arise: the turbulent smoke flow is difficult to model accurately, and the distance between the generator and detector introduces a substantial control-loop delay. This study proposes a smoke simulation method based on model-free adaptive predictive control (MFAPC). The MFAPC scheme was tested in a full-scale aircraft cargo compartment simulator, where it drove the surrogate smoke concentration to track the profile recorded from a real cargo fire. Particle image velocimetry (PIV) was used concurrently with concentration control to capture the corresponding smoke velocity field. Across all conditions, MFAPC reduced the root-mean-square error by up to 38% compared with model-free adaptive control alone. With a control-loop delay longer than 10 s, the light transmission deviation remained within 2% of the target. The PIV data show that the controlled surrogate smoke velocity field reproduces the dominant structures and evolution patterns of actual fire-generated smoke, providing fluid-mechanistic evidence that a recreated dynamic smoke environment is physically meaningful. Full article

The application and performance of Nonlinear Model Predictive Control (NMPC) are critically limited by the computational efficiency of solving nonlinear combinatorial optimization problems. To address this challenge, this study proposes an efficient optimization strategy that employs a neural network to solve the constrained L-1 norm minimization problem within the NMPC framework, thereby enhancing motion control performance. Inspired by the flexible representational capacity and powerful optimization capabilities of neural networks, we explicitly encode the NMPC objective function into a network architecture. The optimal control solution is then obtained efficiently through network training. We validate the proposed strategy in a tractor path-tracking control task, detailing the processes of network construction and optimization. Benefiting from the inherent parallelism and computational efficiency of neural networks, the resulting controller demonstrates excellent real-time performance. Specifically, with prediction horizons set to 5, 10, and 20 steps, the solution times are reduced to less than 0.12, 0.28, and 0.84 s, respectively, under typical operating constraints. Full article

We present ESNformer, a hybrid architecture that couples an Echo State Network (ESN) reservoir with a Transformer encoder for classification of structured, multi-indicator assessment data: a fixed-order vector of complementary indicators per assessment instance rather than a repeated-measures time series. The reservoir acts as a fixed nonlinear feature map over the indicator vector, while self-attention, made position-aware over the fixed column order, learns how each indicator’s evidence contributes to the final decision, so the two components, together, capture local, indicator-level detail and global, cross-indicator interactions within a single, end-to-end trainable model. Interpretability is treated as a first-class design requirement rather than an afterthought: the architecture is paired with an explainability layer combining SHAP feature attribution (reported both globally and per class), the model’s own attention weights, a deletion/insertion faithfulness test that quantitatively verifies which inputs the model actually relies on, and counterfactual maps that translate a prediction into an actionable, inspectable recommendation. We evaluate the architecture on a concrete case study, classifying Braille-literacy instructional recommendations from 15 pedagogical indicators grouped into three categories (Mangold’s, ABKL, and Progresar), using a benchmark of 900 real assessment instances (630 used, together with a class-conditional augmentation procedure, to build a 2100-instance training set) with validation and test partitions (135 instances each) kept exclusively real. On this benchmark, the tuned model reached 85.33% accuracy, 85.90% macro-precision, 85.33% macro-recall, an F1 score of 85.25%, and an AUC of 0.95 on the real test set. SHAP attribution, attention weights, and the faithfulness test converge on the same two dominant indicators (response time and error count): removing them alone collapses accuracy to chance, while retaining only them recovers most of the model’s accuracy. We report this transparently alongside a comparison against ESN-only, Transformer-only, and tabular baselines (logistic regression, decision tree, random forest, XGBoost, and an MLP) on the same data and discuss what the hybrid architecture and its explainability pipeline add beyond what the two dominant indicators already explain and how the approach generalizes to other tabular and mixed-granularity assessment settings that require both predictive accuracy and a verifiable account of what drove each decision. Full article

Real-time monitoring of wave-induced loads supports ship masters’ scientific navigation decisions, where vertical bending moment is a core index representing hull longitudinal bending under waves. This paper combines a temporal convolutional network with the traditional section method to build a vertical bending moment identification model embedded with a section parameter correction mechanism. The main work includes the following: multiple wave condition strain–load datasets are generated via numerical simulations to train the model; the section method calibrates model parameters to boost prediction precision; and wave load tests are conducted to verify the model’s practicability. Results indicate the section method offers physical constraints that embed ship sectional features into the model, lifting identification accuracy, robustness and result rationality. This work applies a temporal convolutional network to the hull vertical bending moment identification with section parameter correction, offering technical references for hull structural safety evaluation and intelligent maritime decision-making. Full article

Background/Objectives: Post-COVID syndrome (PCS) remains a clinically heterogeneous condition. In patients with chronic diseases, real-world evidence is needed to describe its frequency and to identify clinical factors associated with its current presence. This study assessed the prevalence of PCS and factors associated with current PCS status among patients with chronic diseases. Methods: This observational cross-sectional study included a consecutively enrolled source clinical sample of 850 adults receiving inpatient or outpatient care in Almaty and the Almaty Region between June and December 2025. The primary analytic cohort comprised participants with documented prior COVID-19 (n = 250). PCS was assessed at study enrollment according to the national clinical protocol as current symptoms persisting for more than 12 weeks and not explained by an alternative diagnosis. Multivariable logistic regression was used to estimate adjusted odds ratios (ORs) and 95% confidence intervals (CIs). COVID-19 vaccination was excluded from the primary model and evaluated only in a sensitivity analysis because vaccination timing relative to infection was unavailable. The exploratory association-based probability model was evaluated using the AUC with 95% CI, Brier score, calibration assessment, bootstrap internal validation with 1000 resamples, assessment of age nonlinearity, and decision curve analysis. Results: PCS was identified in 147 of 250 participants with prior COVID-19 (58.8%). In the revised primary multivariable model, age (OR = 1.24 per 10-year increase; 95% CI: 1.01–1.54; p = 0.040) and endocrine diseases (OR = 3.36; 95% CI: 1.69–6.98; p < 0.001) remained associated with current PCS status. The association with endocrine diseases persisted after adjustment for body mass index. The apparent AUC was 0.687 (95% CI: 0.620–0.753), and the Brier score was 0.218. Bootstrap internal validation yielded an optimism-corrected AUC of 0.655, a calibration intercept of 0.061, and a calibration slope of 0.816. Conclusions: PCS was frequent among patients with chronic diseases who had prior COVID-19. Older age and endocrine pathology were the most consistent factors associated with current PCS status. All model-performance estimates were obtained within the development sample; the model should not be used for individual prediction or follow-up planning without prospective external validation. Full article

Twin Rotor Multi-Input Multi-Output (TRMS) platforms combine strong aerodynamic cross-coupling, gravitational loading on the vertical axis, friction-dominated horizontal dynamics, and systematic mismatch between idealised models and laboratory hardware. The platform provides only two optical encoders, so the angular rates and the rotor states are unavailable for feedback. This paper presents an observer-based output-feedback architecture that addresses both difficulties. A high-gain observer built on the fully coupled six-state model, including the gyroscopic terms that the control design cannot retain, reconstructs the four unmeasured states from the two encoder angles. The reconstructed states drive a Hybrid Backstepping–Super-Twisting (B-STA) controller in which a second-order continuous sliding mode is embedded at the final recursive step through a composite surface. Because backstepping requires strict-feedback structure, which the centralised coupled model does not possess, the controller is synthesised on a decentralised design model and the residual coupling is rejected as a matched perturbation of the sliding variable. Closed-loop behaviour is analysed as a three-stage cascade covering observer error, sliding variable, and tracking error, yielding practical stability under bounded residuals with an explicit input-to-state gain. The residual bounds are evaluated numerically from the actuator saturation limit and the identified coefficients rather than assumed, and the resulting figures are shown to predict the marked difference in sliding-variable behaviour observed between the two axes. A second architecture applies a windowed Grey Wolf Optimiser (B-GWO) to the backstepping gains, in which each candidate is applied to the plant for a fixed test window, scored on its own accumulated integral of time-weighted absolute error, and followed by a settle window at the incumbent best. We prove that this windowing is a requirement rather than a convenience: a fitness evaluated at a single sample is common to all candidates, cancels from the population ranking, and reduces the search to the minimiser of its own regularisation term. Both schemes are implemented on a physical TRMS through a Simulink Desktop Real-Time interface at a control period of $10\mathrm{ms}$. On a $100\mathrm{s}$ experimental run, B-STA attains a pitch tracking error of $0.0318\mathrm{rad}$ RMS, $7.94\%$ of the reference amplitude, and the lowest control energy on both axes among the strategies compared, reducing pitch control energy by $72.9\%$ relative to a first-order Backstepping–Sliding Mode baseline recorded on the same interface. Numerical disturbance rejection tests on the fully coupled model confirm the mechanism: under a matched actuator step the super-twisting integrator state migrates to a new steady level that cancels the disturbance, driving the residual pitch error to $2\times {10}^{-4}\mathrm{rad}$, whereas the same recursive law without the second-order injection retains a permanent offset of $0.28\mathrm{rad}$. Full article

Background: In high-risk prostate cancer, the optimal timing of radical prostatectomy after diagnostic biopsy remains a clinically important issue. A real-world delay between diagnosis and radical treatment frequently occurs, raising concerns about potential disease progression in high-risk patients. Methods: We conducted a single-center retrospective study including patients with high-risk localized and locally advanced prostate cancer who underwent radical prostatectomy from January 2016 to January 2026. Patients were stratified according to biopsy-to-radical prostatectomy interval into two groups: <90 days or ≥90 days. Adverse pathological outcomes were defined as extraprostatic extension, seminal vesicle involvement, positive surgical margins and lymph-node involvement. Univariable comparisons and multivariable logistic regression analyses were performed to identify independent predictors of adverse pathology. Results: A total of 158 patients with high-risk prostate cancer were included, of whom 67 (42.4%) underwent open- or laparoscopic radical prostatectomy within 90 days after biopsy and 91 (57.6%) after ≥90 days. On univariable analysis, the rates of extraprostatic extension were 59.7% vs. 63.7% in the <90-day and ≥90-day groups (p = 0.606), seminal vesicle involvement was observed in 22.4% vs. 24.2% (p = 0.627), positive surgical margins in 35.8% vs. 39.6% (p = 0.632), and lymph node involvement in 6% vs. 5.5% (p = 0.999). In multivariable logistic regression, a biopsy-to-radical prostatectomy interval ≥90 days was not independently associated with extraprostatic extension (OR 1.38, 95% CI 0.68–2.77, p = 0.371), seminal vesicle involvement (OR 1.43, 95% CI 0.63–3.27, p = 0.391) or positive surgical margins (OR 1.31, 95% CI 0.66–2.62, p = 0.443). The number of positive biopsy cores independently predicted extraprostatic extension (OR 1.17, 95% CI 1.04–1.32, p = 0.008), while higher PSA independently predicted seminal vesicle involvement (OR 1.05, 95% CI 1.01–1.10, p = 0.025) and positive surgical margins (OR 1.05, 95% CI 1.00–1.10, p = 0.027). Conclusions: In this real-world cohort of patients with high-risk prostate cancer undergoing radical prostatectomy, no statistically significant independent association was detected between a biopsy-to-radical prostatectomy interval ≥90 days and extraprostatic extension, seminal vesicle involvement, or positive surgical margins. However, the confidence intervals remained compatible with potentially clinically meaningful differences, and residual confounding from clinician-driven prioritisation and unmeasured preoperative factors cannot be excluded. Therefore, these findings should not be interpreted as demonstrating equivalence or the safety of delaying surgery. Full article

Geothermal energy and underground thermal storage (UTES) are vital to the low-carbon energy transition, yet their optimization is bottlenecked by multi-scale heterogeneity, coupled thermal–hydraulic–mechanical–chemical (THMC) processes, and the high computational cost of full-physics simulations. This review systematically evaluates machine learning (ML) as a foundational paradigm for overcoming these computational and scale-bridging challenges. We categorize current advances into three key functional roles. First, data-driven upscaling directly maps pore-scale features to macro-scale effective properties, replacing traditional empirical homogenization. Second, deep surrogate models mimic high-fidelity THMC simulations at a fraction of the computational cost, enabling real-time prediction and uncertainty quantification. Third, physics-informed digital twins integrate real-time sensor streams with cloud architectures for dynamic reservoir management. Furthermore, we address the generalization limits of purely data-driven approaches, highlighting physics-informed machine learning (PIML) and hybrid architectures that embed conservation laws as strict constraints. Finally, we outline future pathways toward multimodal data fusion and edge-cloud deployment, marking a shift from static offline modeling to dynamic, physics-safeguarded real-time reservoir optimization. Full article

Shrimp farming is a strategic productive activity for Ecuador; however, pond inspection still substantially depends on manual observation and fragmented visual records. This study describes a prototype mobile/web architecture for image capture, storage, and result visualization and, as a separate experiment, evaluates supervised multiclass instance segmentation on public proxy data. A unified dataset of 4508 images was constructed from three external sources using the classes foam, floater, and shrimp. YOLOv8s-seg was used as the internal reference baseline and YOLOv11s-seg as the comparison candidate; both were trained under the same configuration and evaluated on 445 test images. YOLOv8s-seg achieved mAP50 values of 0.678 for BOX and 0.621 for MASK, whereas YOLOv11s-seg achieved 0.677 and 0.616, respectively. Their isolated GPU inference times were 10.4 and 10.2 ms/img. A weighted global experimental performance index comprising 90% predictive quality and 10% inference efficiency reached 0.596 and 0.593, respectively. Both models performed strongly for floater and shrimp, whereas foam showed low recall and zero MASK mAP50 because of the small number of positive test images and heterogeneous annotations. The evaluated task is supervised segmentation of proxy visual categories rather than anomaly detection in its conventional methodological sense, and no general improvement or practical superiority of one architecture was demonstrated. The study does not calibrate an operational confidence threshold or alert-persistence rule, implement adaptive or online learning, validate images from Ecuadorian production ponds, deploy the unified detector within the API, or demonstrate end-to-end real-time monitoring. Consequently, the results constitute a controlled experimental baseline and must not be interpreted as evidence of operational performance or local-domain generalization. Full article

Distributed-drive electric vehicles (DDEVs) enable independent wheel-torque control, providing flexibility to improve energy efficiency and vehicle stability. However, tire–road adhesion, motor and battery capabilities, and actuator availability constrain these objectives, which may conflict under low-adhesion conditions, high-power acceleration, emergency braking, and combined longitudinal–lateral maneuvers. This paper provides a structured review of coordinated torque-allocation strategies for balancing energy efficiency and stability in DDEVs. Existing research is examined in terms of regenerative braking, tire-slip energy-loss reduction, and stability control under longitudinal, yaw, and combined conditions. Control approaches are classified as rule-based, stability-region-based, mode-switching, multi-objective optimization and predictive control, state-adaptive dynamic-priority coordination, and learning-based safety-hybrid methods. These approaches differ in real-time performance, constraint handling, adaptability, interpretability, and engineering maturity. A hierarchical hybrid architecture integrating rule-based supervision, state assessment, constraint-aware optimization, and learning-based enhancement appears more suitable for practical deployment than a single algorithm or fixed-weighting scheme. Key challenges include dynamic stability-boundary estimation, safety-assured coordination, multi-actuator fault tolerance, real-time implementation, and standardized vehicle-level validation. This review provides guidance for coordinated control-system development and future research on DDEVs. Full article

To meet the requirements of differentiated pesticide application between soybean and maize strips in soybean–maize strip intercropping systems, this study developed a strip-specific variable-rate spraying system based on real-time prescription map interpretation and spatiotemporal nozzle matching with delay compensation. A simulated prescription map with predefined application-rate levels was generated using ArcMap and converted into a binary data structure suitable for embedded-controller access. Based on high-precision RTK-BDS positioning information, a local field coordinate transformation model was established and combined with SRAM-based memory preloading to achieve rapid prescription matrix addressing. To reduce boundary misalignment caused by positioning offset, actuator response lag, and hydraulic delay during dynamic field operations, a nozzle spatial position prediction model and a forward delay-compensation control algorithm were developed. Results from the strip-specific variable-rate spraying tests based on the simulated prescription map showed that, under the tested conditions, the mean boundary offset decreased from 0.69 m to 0.29 m after delay compensation. The mean flow-rate control accuracy for both soybean and maize strips exceeded 95% across different application-rate levels, while the coefficients of variation of flow rate were below 6%. These results indicate that, under the tested conditions, the developed system was able to perform real-time prescription map interpretation, target application-rate matching, and strip-specific variable-rate control, demonstrating its technical feasibility for variable-rate spraying in soybean–maize strip intercropping. Full article