Search

The buildup of plaques in atherosclerosis leads to cardiovascular events, with chronic unresolved inflammation and overproduction of reactive oxygen species (ROS) being major drivers of plaque progression. Nanotherapeutics that can resolve inflammation and scavenge ROS have the potential to treat atherosclerosis. Here we demonstrate the potential of black phosphorus nanosheets (BPNSs) as a therapeutic agent for the treatment of atherosclerosis. BPNSs can effectively scavenge a broad spectrum of ROS and suppress atherosclerosis-associated pro-inflammatory cytokine production in lesional macrophages. We also demonstrate ROS-responsive, targeted-peptide-modified BPNS-based carriers for the delivery of resolvin D1 (an inflammation-resolving lipid mediator) to lesional macrophages, which further boosts the anti-atherosclerotic efficacy. The targeted nanotherapeutics not only reduce plaque areas but also substantially improve plaque stability in high-fat-diet-fed apolipoprotein E-deficient mice. This study presents a therapeutic strategy against atherosclerosis, and highlights the potential of BPNS-based therapeutics to treat other inflammatory diseases.

Ring laser gyroscope (RLG) is one of the most important equipment in high-precision strap-down inertial navigation systems and has been widely used in the field of navigation. At present, there is little research on the application of RLG on the angle error measurement of sub-arc-second in the metrology field, and it has not been officially used as the metrology standard instrument. Aiming at the inaccurate measurement of RLG in metrology due to zero position drift and noncoplanar installation, a noncontact high-precision angle measurement error measurement method is proposed in this article. First, the original scale factor is calculated on the standard rotating device to avoid re-calculated on the measured device and solve the traceability problem. Second, the influence of the installation error between the turntable and the laser gyroscope goniometer on the angle measurement error is analyzed, and the compensation method for angle measurement error in the online measurement of the laser gyroscope is proposed. Finally, the feasibility of online metrology of RLG is verified by experiments, and the measurement uncertainty is estimated. The online measurement results show that compared with the photoelectric autocollimator (measurement uncertainty is $U = 0.07^{\prime \prime }({k = 2})$ , the angle measurement error of the RLG goniometer (RLGG) in multiaxis turntable is $ \pm 0.4^{\prime \prime }$ , and the measurement uncertainty is $U = 0.36^{\prime \prime }({k = 2})$ .

Endowing quadruped robots with the skill to forward jump is conducive to making it overcome barriers and pass through complex terrains. In this paper, a model-free control architecture with target-guided policy optimization and deep reinforcement learning (DRL) for quadruped robot jumping is presented. First, the jumping phase is divided into take-off and flight-landing phases, and optimal strategies with solt actor-critic (SAC) are constructed for the two phases respectively. Second, policy learning including expectations, penalties in the overall jumping process, and extrinsic excitations is designed. Corresponding policies and constraints are all provided for successful take-off, excellent flight attitude and stable standing after landing. In order to avoid low efficiency of random exploration, a curiosity module is introduced as extrinsic rewards to solve this problem. Additionally, the target-guided module encourages the robot explore closer and closer to desired jumping target. Simulation results indicate that the quadruped robot can realize completed forward jumping locomotion with good horizontal and vertical distances, as well as excellent motion attitudes.

The pathogenesis of cancer and cardiovascular diseases is subjected to spatiotemporal regulation by the tissue microenvironment. Multiplex visualization of the microenvironmental components, including immune cells, vasculature and tissue hypoxia, provides critical information underlying the disease progression and therapy resistance, which is often limited by imaging depth and resolution in large-volume tissues. To this end, light sheet fluorescence microscopy, following tissue clarification and immunostaining, may generate three-dimensional high-resolution images at a whole-organ level. Here we provide a detailed description of light sheet fluorescence microscopy imaging analysis of immune cell composition, vascularization, tissue perfusion and hypoxia in mouse normal brains and hearts, as well as brain tumors. We describe a procedure for visualizing tissue vascularization, perfusion and hypoxia with a transgenic vascular labeling system. We provide the procedures for tissue collection, tissue semi-clearing and immunostaining. We further describe standard methods for analyzing tissue immunity and vascularity. We anticipate that this method will facilitate the spatial illustration of structure and function of the tissue microenvironmental components in cancer and cardiovascular diseases. The procedure requires 1–2 weeks and can be performed by users with expertise in general molecular biology.

Deep-learning-based detection methods have been widely applied to industrial defect inspection. However, directly using vanilla detection methods fails to achieve satisfying performance due to the lack of identifiable features. In this article, a novel attention-based multiscale feature fusion module (AMFF) is proposed, aiming to enhance defect features and improve defect identification by leveraging attention mechanism in the feature fusion. AMFF includes self-enhanced attention module (SEAM) and cross-enhanced attention module (CEAM). SEAM is performed on a single feature map, which first adopts multiple dilation convolutions to enrich contextual information without compromising resolution and then utilizes the intralayer attention on the current feature map. CEAM takes both the current feature map and the adjacent feature map as input to perform cross-layer attention. The adjacent feature map is modulated with the guidance of the current feature map, which is then combined with the current feature map and the output of SEAM for final prediction. AMFF is utilized in current feature fusion networks, e.g., feature pyramid network (FPN) and path aggregation FPN (PAFPN), and is further integrated into prevalent detectors to guide them to pay more attention to defects rather than the background. Extensive experiments are conducted on two real industrial datasets released by Tianchi platform, i.e., fabric and aluminum defect datasets. For each dataset, 500 images are randomly selected for test and the rest for training. The proposed AMFF is demonstrated to significantly boost defect detection accuracy with acceptable computational cost, and the real-time performance could fully satisfy practical requirements.

This article presents a study on trajectory tracking control for an underactuated bionic underwater vehicle that is propelled by undulating fins. The vehicle's propulsion mechanism is inspired by marine organisms, and its working principle is briefly introduced. Accurate trajectory tracking control is challenging due to the coupling of fin motions and the lack of control degree of freedom. The multi-input and multi-output characteristics of the tracking control task, coupled with the existence of system constraints, impose high requirements for the controller. To solve these problems, a novel dynamics model-based predictive control strategy is proposed and the control output at the dynamics level is obtained. To stabilize the underactuated dynamics, a practical sidesway compensator is designed and integrated into the control framework. The convergence of the approach is guaranteed. Real-time precise tracking control is realized. Extensive simulations and real-world experiments are carried out and compared with other control methods. Reliability and superiority of the proposed approach is demonstrated with up to an 80% reduction in trajectory tracking sidesway errors.

The transformation of metastable ferrihydrite to stable hematite has been linked to magnetic enhancement in soils and sediments. However, the magnetic and structural characteristics associated with the transformation of arsenate (As(v))-coprecipitated ferrihydrite to hematite remain unclear. Herein, we have investigated the structural and magnetic characteristics of fresh and aged As(v)-ferrihydrite coprecipitates. Magnetic characterization indicates that the high As(v) loading weakens the ferrimagnetic susceptibility of fresh Fe–As(v) coprecipitates by forming Fe–AsO4complexes. In the low As(v)-loading system (Fe/As(v) ≥ 40), the increase in As(v) concentrations accelerates the growth of hematite particles with an expanded unit cell parameter c, and enhances magnetic susceptibility by changing both the surficial and internal lattice arrangements of hematite crystallites. In contrast, in the high As(v)-loading system (Fe/As(v) ≤ 20), the transformation of ferrihydrite is greatly retarded and the formation of poorly crystalline ferrimagnetic maghemite significantly enhances the magnetic susceptibility in the transformation products. Our findings provide new insights into the enhancement of magnetic properties of iron (oxyhydr)oxides in high arsenic soils and sediments, which are also essential for developing effective immobilization and long-term storage strategies for arsenic using iron (oxyhydr)oxides.

Disturbances caused by unknown dynamics and environmental factors render the automatic control of underwater vehicles extremely challenging. These effects are complex, time varying, and difficult to model accurately, leading to possible instability in the control process. This article focuses on the disturbance rejection problem of an underactuated bionic underwater vehicle (BUV). A novel control approach that combines the robustness of a model predictive controller with the modeling capabilities of a learning-based observer is proposed. More specifically, a practical Gaussian process (GP)-based disturbance observer is designed for online disturbance estimation. The interactive prediction results are integrated into a model predictive controller to obtain a refined dynamics model. The control framework optimally solves for the optimal control output to achieve control at the dynamics level. The proposed approach realizes disturbance rejection control while ensuring real-time application, constraint satisfaction, and recursive feasibility under complicated disturbances. The feasibility and efficiency of the proposed approach are verified through simulations and real-world experiments.

Thickness is a crucial parameter that significantly influences the optical, piezoelectric, and magnetoelectric properties of films. Reducing the film’s thickness nonuniformity at wafer scale is quite important for industrial manufacturing but difficult and challenging. Conventionally, the complementary metal-oxide-semiconductor transistor (CMOS)-compatible magnetron sputtering (MS) method employs planar targets, which often leads to an unsatisfactory thickness distribution, i.e., thick in the center and thin at the edges. To address this issue, a concave target is proposed based on the theoretical thickness distribution model and is experimentally manufactured to optimize the thickness nonuniformity of magnetron-sputtered films. Additionally, the influences of target size and target–substrate distance are also investigated. Our results demonstrate that using a concave target with a large radius is effective for improving the thickness uniformity. For instance, ultrauniform Mo films are deposited onto a 2-in. Si/SiO2wafer, exhibiting a notably low thickness nonuniformity of 0.38%. Our results provide theoretical guidance and the experimental prototype for the large-scale preparation of ultrauniform films. Moreover, these findings could be applied to other materials.

Light availability for phytoplankton in shallow lakes is closely related to non‐algal light attenuation (KdNALA, the fractional light absorption by non‐algal substances, such as suspended sediments); thus, significant changes in global wind speed in recent decades may have a profound effect on light availability and algal blooms in shallow lakes. Herein, the eutrophic shallow Lake Taihu was selected to investigate the long‐term dynamics of light availability and its effect on algal blooms. The results showed that KdNALAdecreased from approximately 3.5 to 2.5 m−1with decreasing wind speed from 2005 to 2021, indicating the significantly increased light availability for phytoplankton despite surface photosynthetically active radiation exhibiting limited variability during the study period. In addition, both experimental (i.e., in situ nutrient enrichment experiments) and statistical approaches (i.e., deviations of Trophic State Index subindices) indicated that phytoplankton growth was primarily light‐limited during the study period in Lake Taihu. Consequently, considerable increases in fraction of observed and maximal chlorophyll ayield at given nitrogen or phosphorus concentration were observed and were mostly related to decreasing KdNALA, which implied that decreasing KdNALAallowed phytoplankton utilize “unused nutrient‐capacity” until the additional algal‐turbidity induce further light limitation or nutrient limitation. As the effect of changes in global wind speed on KdNALAand algal growth received limited attention in the existing research, we revealed an underappreciated mechanism by which global changes in wind speed significantly affects algal biomass by influencing light availability, which may have profound effects on future algal bloom mitigation efforts in shallow lakes.

Urban sewage sludge treatment is important for sustainable utilisation and virtuous cycle of freshwater resources. However, with the improvement of sewage discharge standards, ensuring stable operation of sewage sludge treatment plants is becoming an urgent problem to be solved in the sewage treatment industry. This paper proposes a FNN control framework based on different working conditions to optimise the whole process of municipal sewage sludge treatment and discharge. The framework first divides the working conditions according to the weather, forming a separate feature and an input vector together with the typical indicators of other sewage treatment plants. Then the FNN is used to complete the control and optimisation of various indicators, achieving the dual objectives of reducing energy consumption and optimising water quality. Finally, the model is tested for the tracking index of sewage flow. The results demonstrate that the FNN control method used has significantly lower MAE than the single method in the two indexes of energy consumption and water quality evaluation. This provides new ideas for the optimisation of urban sewage sludge treatment process in the future. Overall, the paper effectively highlights the importance of urban sewage sludge treatment and presents a well-designed FNN control framework for optimising the treatment process. Additionally, the paper could benefit from further elaboration on the significance of the results obtained, and suggestions for future research in this area.

Moving object segmentation (MOS), aiming at segmenting moving objects from video frames, is an important and challenging task in computer vision and with various applications. With the development of deep learning (DL), MOS has also entered the era of deep models toward spatiotemporal feature learning. This paper aims to provide the latest review of recent DL-based MOS methods proposed during the past three years. Specifically, we present a more up-to-date categorization based on model characteristics, then compare and discuss each category from feature learning (FL), and model training and evaluation perspectives. For FL, the methods reviewed are divided into three types: spatial FL, temporal FL, and spatiotemporal FL, then analyzed from input and model architectures aspects, three input types, and four typical preprocessing subnetworks are summarized. In terms of training, we discuss ideas for enhancing model transferability. In terms of evaluation, based on a previous categorization of scene dependent evaluation and scene independent evaluation, and combined with whether used videos are recorded with static or moving cameras, we further provide four subdivided evaluation setups and analyze that of reviewed methods. We also show performance comparisons of some reviewed MOS methods and analyze the advantages and disadvantages of reviewed MOS methods in terms of technology. Finally, based on the above comparisons and discussions, we present research prospects and future directions.