此英文版本为编辑部提供。The English version is provided by the editorial office. This paper investigates the event-triggered adaptive inverse optimal control problem for a class of stochastic multiple-input multiple-output (MIMO) nonlinear cyber-physical systems (CPSs) under hybrid cyber-attacks. The hybrid cyber-attack model, consisting of denial-of-service (DoS) and false data injection (FDI) attacks, is characterized by a probabilistic Bernoulli process to better capture realistic attack scenarios. Based on the backstepping approach, an event-triggered adaptive optimal control scheme is developed by integrating inverse optimal design with a switching-threshold event-triggered mechanism. The proposed method is applicable to stochastic MIMO systems and ensures that the closed-loop system is semi-globally uniformly ultimately bounded (SGUUB) in probability. Finally, simulation results are presented to demonstrate the effectiveness and feasibility of the proposed strategy.
此英文版本为编辑部提供。The English version is provided by the editorial office. To establish a tumor-endothelial cell co-culture model with a spatially partitioned structure, this study employed a chitosan electrostatic adsorption coating strategy to surface-modify calcium alginate microspheres. Breast tumor cells (MCF-7) were encapsulated within the microspheres, while human umbilical vein endothelial cells (HUVECs) were seeded on the microsphere surface, mimicking the spatial relationship between tumor cell clusters and vascular endothelium in solid tumors. By systematically investigating the effects of varying chitosan mass concentrations (1, 3, 5, 7 g/L) on microsphere particle size, surface charge, dimensional stability, and HUVECs adhesion, the optimal coating conditions were identified, and the proliferation behavior of tumor cells in the co-culture model was evaluated. Results indicated that microspheres modified with 5 g/L chitosan exhibited optimal particle size stability, a surface zeta potential of (9.00±0.57) mV, the highest HUVECs adhesion density with good spreading, and no significant cytotoxicity toward internal MCF-7 cells. In the co-culture model constructed using this optimized coating, HUVECs formed a continuous cell layer on the microsphere surface, connecting dispersed microspheres into tissue-like aggregates. MCF-7 cell activity in the co-culture group was enhanced compared to the monoculture group (reaching 115% on day 7). This model showed potential applications in studying tumor-endothelial cell interaction mechanisms, drug evaluation, and personalized medicine, providing a novel in vitro platform for simulating the tumor microenvironment.
此英文版本为编辑部提供。The English version is provided by the editorial office. A safe reinforcement learning method is proposed for gasoline blending scheduling under random disturbances in the properties and flow rates of direct-feed component oils. The scheduling task requires the coordinated optimization of discrete scheduling decisions and continuous blending recipes under strict product quality and inventory constraints. To solve this problem, the scheduling process is formulated in a hybrid action space so that discrete product switching decisions and continuous recipe regulation can be handled in a unified framework. On this basis, a Multi-pass hybrid soft actor-critic algorithm (MPH-SAC) is developed for gasoline blending scheduling. The Multi-pass structure removes spurious gradients in hybrid action evaluation and improves the coordination between discrete scheduling decisions and continuous blending recipes. In addition, a proportional-integral-derivative (PID) regulated Lagrangian multiplier mechanism is introduced to dynamically enforce safety constraints while maintaining economic performance during learning. The proposed method is tested in a gasoline blending scheduling environment with uncertain directfeed component oil conditions. Simulation results show that the proposed method achieves zero violations of product quality and inventory constraints under random disturbances. The average online decision time is about 5 ms, whereas the deterministic optimization model solved by Gurobi requires about 394 s to obtain a near-optimal solution. Although the average profit of the proposed method is about 2.5% lower than the deterministic optimum, it maintains strong robustness against disturbances in the properties and flow rates of direct-feed component oils and avoids the model mismatch caused by deterministic assumptions. These results demonstrate that the proposed method can effectively balance profitability, operational safety, and real-time decision requirements, and it provides a practical solution for real-time gasoline blending scheduling in complex and uncertain operating environments.
此英文版本为编辑部提供。The English version is provided by the editorial office. To address the issues of high cost and strong subjectivity associated with manual observation in dairy cow estrus detection, as well as the high computational power consumption caused by high-dimensional model features, this study proposes a dairy cow estrus detection method based on unsupervised clustering and interpretable feature selection. The method utilizes the K-means algorithm to cluster raw activity time series data collected by neck-mounted accelerometers on dairy cows to automatically generate estrus behavior labels. Three models are compared: K-Nearest Neighbors (KNN), Support Vector Machine (SVM), and Random Forest (RF). These models are evaluated under time windows of 10, 30 s, and 50 s. Finally, the optimal model is determined by the Friedman test and the Nemenyi post-hoc test. The Shapley Additive Explanations (SHAP) method is introduced to quantify feature contributions and screen for the key features. The results indicate that the RF model based on a 10 s time window achieves the best performance. Through SHAP selection, the features are reduced from 25 to 3. The performance of the dimension-reduced model shows no significant deviation, with accuracy, precision, recall, and F1 score reaching 0.991 1, 0.937 0, 0.995 5, and 0.965 4, respectively. In independent trials based on three-axis acceleration data from dairy cows, the RF model achieves an estrus detection accuracy of 0.909 0 using a 10 s window following dimensionality reduction. The proposed method enables both estrus label generation and feature dimensionality reduction, while maintaining model interpretability. This provides technical support for developing low-power, long-endurance estrus monitoring devices for dairy cows.
此英文版本为编辑部提供。The English version is provided by the editorial office. Retrosynthetic analysis is the process of deriving commercially available precursor molecules from target product molecules. To address the limitations of existing single-step chemical retrosynthetic methods in terms of generalization ability and prediction accuracy, this paper proposes an ensemble prediction model integrating molecular fingerprints and graph neural networks, which addresses the poor scalability and interpretability of individual models. This method improves the accuracy and stability of retrosynthetic prediction by constructing multidimensional molecular representations and dynamically selecting the optimal prediction strategy. The model is based on messagepassing neural networks (MPNNs), global self-attention mechanisms, and AttentiveFP networks to extract molecular structural features, and introduces extended connectivity fingerprints (ECFP) to achieve joint modeling of local structural information and global molecular features, constructing two fusion architectures: MPNN+FPS (Fingerprints)and Attentive FP+FPS. Extensive experiments on the USPTO-50K and natural-product datasets validate the effectiveness of the method. The proposed ensemble retrosynthetic prediction method demonstrates significant advantages in both prediction accuracy and generalization ability, providing a new technical approach for retrosynthetic analysis of complex molecules and natural products.
此英文版本为编辑部提供。The English version is provided by the editorial office. This paper proposes an optimal event-triggering control method based on prescribed performance constraints for nonlinear zero-sum game systems with unknown dynamics, achieving prescribed performance control for systems with unknown dynamics under reduced computational resources. First, by introducing a prescribed performance function, the constrained system control problem is transformed into an equivalent unconstrained optimal control problem. Second, leveraging an off-policy data iteration mechanism, combined with polynomial fitting techniques and the Integral Reinforcement Learning (IRL) method, a model-free iterative control strategy is derived within a purely data-driven control law framework, addressing the challenges posed by unknown system dynamics. Furthermore, an event-triggering mechanism based on game theory is designed, effectively reducing the number of samples and computational burden by modeling event-triggering sampling errors and control strategies as a zero-sum game. Then, a single critic neural network is employed to approximate the optimal control solution, with the proposed strategy used to update the network weights. Finally, theoretical analysis and simulation results demonstrate the convergence and effectiveness of the proposed method.
此英文版本为编辑部提供。The English version is provided by the editorial office. Major Depressive Disorder (MDD) is a mental illness that severely affects human health. Currently, depression recognition based on Contrastive Learning (CL) for Electroencephalogram (EEG) signals has achieved promising results, yet it still has certain limitations: Mainstream CL methods primarily focus on the temporal consistency of EEG signals while neglecting the topological consistency of EEG data, namely the anti-interference stability of individual channels and the inherent neurophysiological correlations between channels. To address the above issues, this paper proposes a novel depression recognition model based on Temporal-Topological Consistency Graph Contrastive Learning (TTGCL). First, an adaptive graph augmentation strategy is designed, which includes dynamic node augmentation based on signal-to-noise ratio and frequency-domain energy distribution, as well as edge augmentation based on dynamic regular constraints, to generate high-quality contrastive views. Second, Cross-Window Temporal Consistency Contrast (CWTCC) is proposed to learn the temporal consistency of EEG signals through crossprediction between dual views. Finally, a Regional-Global Graph Contrast (RGGC) framework is constructed to learn topological consistency from the level of local brain regions to global samples. This study achieves recognition accuracies of 96.53% and 96.33% on two public MDD datasets, respectively, outperforming existing baselines.
此英文版本为编辑部提供。The English version is provided by the editorial office. To investigate the suppression effect of porous materials on turbulent interaction noise at the leading edge of airfoils and reveal the noise reduction mechanism, numerical simulations of the classic rod-airfoil (NACA0012)interaction noise model were conducted using Large Eddy Simulation (LES) combined with the FW-H acoustic analogy method. The numerical setup was validated by comparison with experimental data and grid sensitivity analysis. This study compared the flow field and sound field characteristics of a solid leading edge and porous leading edges with porosities of 40%, 50%, and 60%, respectively. The results showed that porous leading edges could effectively attenuate low-and mid-frequency single-tone noise, and the noise reduction performance improved as porosity increased. A maximum noise reduction of 11.2 dB was achieved, yet the average drag also rose correspondingly. Analyses of pressure fluctuations, velocity distributions within the stagnation region, and turbulent coherence were also performed. The results indicated that the porous medium suppressed the vertical velocity component of the incoming flow as the flow impinged on the leading edge; consequently, the pressure fluctuations were also suppressed. Furthermore, the porous structure disrupted the spanwise coherence of turbulent structures and weakened acoustic radiation efficiency.
此英文版本为编辑部提供。The English version is provided by the editorial office. With the rapid development of AI + medicine technology and a significant increase in corresponding patent applications, international competition is fierce, and China’s AI + medicine patent examination is facing new challenges. This article provides an in-depth analysis of the current status of AI + medicine patent applications and the problems in inventive examination. It is proposed to use the technical problems actually solved by the invention as the core, and reasonably evaluate creativity while fully considering the technical characteristics of the pharmaceutical field. From the perspectives of evaluating the value of input data, selecting and improving AI tools, and examining experimental data, specific cases are analyzed to provide support for improving the quality of AI + medicine patents in China.
此英文版本为编辑部提供。The English version is provided by the editorial office. Pertussis is a respiratory infectious disease caused by Bordetella pertussis (B. pertussis). Despite the currently high vaccination coverage, a considerable number of pertussis cases are still reported every year.Various explanations have been proposed for the recurrence of pertussis, among which the adaptive evolution of B. pertussis is considered a major factor. With the isolation of an increasing number of circulating strains, strain typing has revealed that the genotypes of circulating strains have gradually diverged from those of vaccine strains and have formed new lineages. This is mainly reflected in the emergence of ptxP3 strains with enhanced virulence, strains that lack vaccine antigen components to escape vaccine induced immunity, and strains resistant to macrolide antibiotics. The use of pertussis vaccine has, to some extent, promoted the adaptive evolution of B. pertussis. The evolved strains have influenced the preventive efficacy of current vaccines. This review summarizes four aspects: the current epidemiological status of circulating strains, the microevolution of B. pertussis, the association between pertussis vaccines and the microevolution of circulating strains, and the protective efficacy of pertussis vaccine. It aims to clarify the relationship between the evolutionary direction of circulating strains and pertussis vaccines and therapeutic drugs, to evaluate the protective efficacy of current vaccines against circulating strains, and to provide a reference for studies on the evolutionary direction, epidemic trends, and prevention and treatment of pertussis in China.