Abstract:The selective absorption and scattering effects of water on visible light of different wavelengths severely constrain the quality of optical imaging. Polarization imaging technology demonstrates significant advantages in high-turbidity, short-range underwater environments. This study employs an active polarization imaging system utilizing three laser wavelengths—red, green, and blue—to comparatively analyze imaging performance in typical natural water with turbidity levels of 10-25 NTU. Experimental results indicate that red light achieves the optimal polarization imaging performance, followed by green light, while blue light performs the weakest. Furthermore, a polarization image enhancement algorithm is proposed, which significantly improves the imaging quality for all three laser wavelengths. Under a turbidity condition of 19.97 NTU, the entropy value of enhanced images of diving suit fabric using the proposed method shows an approximately 34.4% improvement compared to the traditional Schechner method. The research demonstrates that integrating polarization imaging into active laser imaging systems can effectively enhance underwater imaging quality, offering new insights for optimizing underwater imaging technology at specific wavelengths.

Abstract:G protein-coupled receptors (GPCRs) constitute a crucial superfamily of membrane proteins that play a pivotal role in cellular signal transduction and serve as primary targets in contemporary drug development. The β2-adrenergic receptor (β2AR), a representative member of class A GPCRs, is a critical target in the therapeutic management of respiratory diseases. Despite the availability of several β2AR agonists in clinical practice, there remains a substantial need for optimization concerning drug safety, efficacy, and receptor selectivity. In this study, a virtual screening approach was utilized to effectively identify β2AR agonists from a compound library comprising 19 million molecules. Through comprehensive cellular assays and in vivo pharmacokinetic evaluations, a novel short-acting agonist with an EC50 value of 0.86 nM was discovered, presenting a promising candidate for the development of next-generation treatments for respiratory diseases.

Abstract:Currently, AI application workloads, represented by machine learning, exhibit a dual-density characteristic, combining both compute-intensive and data-intensive traits. These applications not only require support for the storage, transmission, and fault tolerance of massive data but also need to optimize the performance of complex logical computations. Traditional single big data frameworks or high-performance computing frameworks can no longer meet the challenges posed by these applications. The hybrid big data platform based on Spark and MPI proposed in this paper is a high-performance big data processing platform. This platform, built on a typical large-scale cluster, focuses on addressing the storage and computing characteristics of dual-density applications, such as those in machine learning, and includes three key modules: dual-paradigm hybrid computation, heterogeneous storage, and integrated high-performance communication. To address the dual-density nature of these applications, which involve both data-intensive big data processing and compute-intensive high-performance computing, a computational module combining the Spark and MPI paradigms is designed. By splitting and classifying tasks, compute-intensive tasks are offloaded to the MPI computation module, enhancing the dual-paradigm hybrid computation capability. To address the different data characteristics during the computation process, a heterogeneous storage structure and a data-metadata separation strategy are designed. This optimizes data storage through classification, building a highly available, high-performance storage system. In response to the communication needs of dual-density computing, a high-performance communication technique is proposed, providing strong communication support for the computing and storage modules. Test results show that this platform provides efficient dual-paradigm hybrid computation for dual-density applications, achieving performance improvements of 4.2% to 17.3% compared to a standalone Spark big data platform for various computation tasks.

Abstract:Targeted protein degradation technologies face significant challenges, including insufficient target specificity and low delivery efficiency. Mesoporous bioactive glass (MBG) is a biocompatible nanomaterial widely studied in drug delivery, yet its potential as a platform for targeted protein degradation remains unexplored. In this study, FITC labeling and the biotin-avidin system were employed to evaluate the subcellular localization of MBG and its potential as a protein degradation carrier. Additionally, the ferroptosis-inducing capability of Fe-doped MBG was investigated. The results demonstrated that MBG facilitates the internalization of target proteins and degradation in lysosomes, as exemplified by the degradation of PD-L1. Furthermore, MBG was shown to deliver iron ions into lysosomes, inducing ferroptosis. This study, for the first time, reveals the dual functionality of MBG in targeted protein degradation and ferroptosis induction, offering an innovative approach for the spatiotemporally controlled synergistic treatment of cancer through "protein degradation-ferroptosis" strategies.

Abstract:In order to better monitor the health state of power lithium batteries. A lithium battery health state prediction method based on Improved Bidirectional Temporal Convolutional Network, Long Short Term Memory Network and Attention Mechanism is proposed. The hyperparameters of the proposed method are optimized using Crested Porcupine Optimizer. Tests were conducted on the University of Maryland lithium battery charge/discharge dataset to extract capacity-related health features, and the health features with higher correlation were screened by Pearson correlation coefficient as inputs to the neural network algorithm. The Root Mean Squard Error of the proposed method is no more than 0.020, the Mean Absolute Error is no more than 0.017, and the R-Square is above 0.995 for all battery health state predictions. Higher accuracy can be achieved in lithium battery health state prediction.

Abstract:As the internal electrode materials in multilayer ceramic capacitors (MLCCs) are increasingly being replaced by base metals such as nickel instead of precious metals, the sintering process must be conducted in a reducing atmosphere. This work investigates the properties of Mn-doped BaTiO3-based ceramics sintered under different reducing atmospheres, exploring the effect of varying H?/N? ratios on dielectric performance and reliability. BaTiO3-based ceramics were prepared using the solid-phase method under different reducing atmospheres. X-ray diffraction (XRD), scanning electron microscopy (SEM), Raman spectroscopy, and X-ray photoelectron spectroscopy (XPS) were employed to characterize the microstructure and properties of the samples. The study found that the polarization mechanism significantly affects the dielectric performance and reliability of BaTiO3-based ceramics. Specifically, as the reducing atmosphere intensifies, the polarization mechanism transitions from short-range defect dipole polarization to long-range defect carrier polarization, impacting the dielectric constant, dielectric loss, and insulation resistance. The sample sintered under 1.5% H?/98.5% N? (S2) exhibited the highest dielectric constant, lowest dielectric loss, and best insulation resistance, demonstrating excellent overall performance. Additionally, BaTiO3-based MLCCs with thickness of 0.9 μm and the TCC of X7R were successfully fabricated. This work provides theoretical and technical guidance for improving the dielectric performance and reliability of BME-MLCCs.

Abstract:3D scene reconstruction is a critical research topic in autonomous driving, robotics, and related fields, with extensive applications in navigation mapping, environmental interaction, and virtual/augmented reality tasks. Current deep learning-based reconstruction methods can be primarily categorized into five groups from the perspectives of scene representation and core modeling techniques: cost volume-based depth estimation methods, truncated signed distance function (TSDF)-based voxel approaches, transformer architecture-based large-scale feedforward methods, multilayer perceptron (MLP)-based neural radiance fields (NeRF), and 3D Gaussian Splatting (3DGS). Each category exhibits unique strengths and limitations. The emerging 3DGS method distinguishes itself by explicitly representing scenes through Gaussian functions while achieving rapid scene rendering and novel view synthesis via efficient rasterization operations. Its most significant advantage lies in diverging from NeRF''s MLP-based scene representation paradigm - 3DGS ensures both efficient rendering and interpretable editable scene modeling, thereby paving the way for accurate 3D scene reconstruction. However, 3DGS still faces numerous challenges in practical scene reconstruction applications. Based on this analysis, this paper first provides a concise introduction to 3DGS fundamentals and conducts comparative analysis with the aforementioned four categories. Following a systematic survey of existing 3DGS reconstruction algorithms, we summarize the key challenges addressed by these methods and review current research progress on core technical difficulties through representative case studies. Finally, we prospect potential future research directions worthy of exploration.

Abstract:The study investigates the influence of different raw material particle sizes of barium titanate (BaTiO?, BT) on the dielectric properties of ceramics and aims to optimize their electrical performance through doping modifications. Ceramic samples were prepared using BT powders with particle sizes of 100 nm, 150 nm, 200 nm, and 250 nm via the solid-state ball milling method. Y?O?, Ho?O?, MgO, and SiO? were introduced as dopants to control grain growth and tailor the dielectric properties. The microstructure and dielectric characteristics of the ceramic samples were systematically analyzed using scanning electron microscopy (SEM), transmission electron microscopy (TEM), X-ray diffraction (XRD), and dielectric measurements. The results revealed that the BT-10 sample exhibited significant dopant diffusion, which hindered the formation of an ideal "core-shell" structure and reduced tetragonality. While BT-10 showed a relatively high dielectric constant, its temperature stability was poor. In contrast, the BT-15, BT-20, and BT-25 samples successfully formed a "core-shell" structure, with BT-25 demonstrating the highest tetragonality. The BT-25 sample exhibited the most favorable electrical performance, including the highest saturation polarization strength (Ps = 11.817 μC/cm2), remanent polarization strength (Pr = 1.465 μC/cm2), and superior dielectric stability under DC bias conditions. These findings indicate that the BT-25 sample offers the most balanced and optimized overall performance. Furthermore, the study highlights the challenges associated with using smaller BT particle sizes in fabricating "core-shell" structured ceramics, particularly the increased specific surface area and higher defect density of BT powders. This research provides valuable theoretical insights and technical guidance for optimizing the dielectric layers of multilayer ceramic capacitors (MLCCs).

Abstract:Hand pose estimation based on RGB images shows crucial application prospects in the fields of dynamic gesture recognition and human-computer interaction. However, existing methods face numerous challenges. For example, the high degree of self-similarity of the hand and the extremely dense distribution of key points make it extremely difficult to achieve high-precision prediction under the condition of low computational cost, which in turn leads to limitations in performance in complex scenarios.In view of this, this paper proposes a two-dimensional (2D) hand pose estimation model based on the YOLOv8 network, namely FAR-HandNet. This model ingeniously integrates the Focused Linear Attention module, the key point alignment strategy, and the regression residual fitting module, effectively enhancing the feature capture ability for small target areas (such as the hand), while reducing the adverse impact of self-similarity on the positioning accuracy of hand key points. It is worth mentioning that the regression residual fitting module uses a flow-based generative model to fit the distribution of key point residuals, which greatly improves the accuracy of the regression model.The experiments in this paper are carried out on the CMU and FreiHAND datasets. The experimental results clearly show that FAR-HandNet has obvious advantages in terms of the number of parameters and computational efficiency, and performs excellently in PCK (Percentage of Correct Keypoints) under different thresholds, showing a significant improvement compared with existing methods. In addition, the inference time of this model is only 32ms. The ablation experiments further confirm the effectiveness of each module, fully verifying the effectiveness and superiority of FAR-HandNet in the hand pose estimation task.

Abstract:The loss of upper limb function brings a lot of inconvenience to the life of amputees. In order to improve the life quality of upper limb amputees, it is necessary to develop a low-cost, lightweight and powerful prosthetic system. In this paper, a three-degree-of-freedom modular light-weight upper limb prosthetic arm and its multi-joint cooperative control system are designed to provide a lightweight, economical, modular and comprehensive prosthetic solution. By using the hollow structure design, the overall weight of the prosthetic arm is significantly reduced (about 2 kg), which is much lower than existing commercial prosthetic products, while ensuring the number of degrees of freedom and effectively reducing manufacturing costs, improving the comfort and suitability of the prosthetic arm. In addition, the multi-joint control system designed in this paper, combined with the precise coordination algorithm, can accurately control each joint to reach a predetermined Angle at the same time, to meet the needs of different degrees of amputees for multi-joint cooperative movement. Through precision and efficiency tests, the results show that the prosthesis performs well in terms of control accuracy, and the motion efficiency can meet most of the needs of daily life.