Search

For wearable smart textile sensors, stability, accuracy and multi-functionality are key objectives. Achieving the optimal application requires delicately balancing the crucial physical properties of strain sensors, presenting a key technological challenge. This study addresses these challenges by presenting several properties and potential applications of a triple hierarchic polymeric knitted fabric. The fabric incorporates an internal conductive network constructed with silver nanowires (AgNWs) and polydopamine (PDA) coating on its outer surface. This innovative textile successfully strikes a balance between strain sensing and electromagnetic interference shielding while concurrently exhibiting biocompatibility and antimicrobial properties. Significantly, acknowledging the susceptibility of measurements from polymer-based strain sensor materials to time drift, we introduce both a modeling approach and a novel calibration technique. This advancement facilitates the generation of stable cyclic sensing signals, even under substantial deformations of up to 80 % at a high stretching speed. Importantly, it provides a practical solution for addressing signal drift observed in flexible sensors when utilized in environments characterized by long-term and large deformations., QC 20240612

Wing sail-assisted technology is an option for the cargo ship retrofitted or designed to meet the requirement of EEXI or EEDI. In this paper, the factors influencing the EEDI contribution of wing sails for a VLCC have been carried out based on the wind tunnel and tank model test results, including the arrangements of sails, the total area of sails, the sailing speeds of the ship, and wind probabilities along the main global routes. Investigation of the influence of the side force induced by the sails into the net EEDI contribution was also provided considering drift and rudder angle variation of the ship. The conclusions obtained in this study show that wind probability is a sensitive factor, and when following the method in MEPC.1/Circ.896, the EEDI contribution of four sails for the VLCC can reach almost 6%. The EEDI contribution of the sails tends to increase as the ship speed decreases. The total area of the sails has a greater effect on the EEDI contribution than the sail arrangement. Wing sails' contribution to EEDI will be overestimated by the method described in MEPC.1/Circ.896, which ignores variations in the drift and rudder angle of the ship, compared to MMG-based model results.

End-to-end autonomous driving has witnessed remarkable progress. However, the extensive deployment of autonomous vehicles has yet to be realized, primarily due to 1) inefficient multi-modal environment perception: how to integrate data from multi-modal sensors more efficiently; 2) non-human-like scene understanding: how to effectively locate and predict critical risky agents in traffic scenarios like an experienced driver. To overcome these challenges, in this paper, we propose a Multi-Modal fusion transformer incorporating Driver Attention (M2DA) for autonomous driving. To better fuse multi-modal data and achieve higher alignment between different modalities, a novel Lidar-Vision-Attention-based Fusion (LVAFusion) module is proposed. By incorporating driver attention, we empower the human-like scene understanding ability to autonomous vehicles to identify crucial areas within complex scenarios precisely and ensure safety. We conduct experiments on the CARLA simulator and achieve state-of-the-art performance with less data in closed-loop benchmarks. Source codes are available at https://anonymous.4open.science/r/M2DA-4772.

Integrating LiDAR and camera information into Bird's-Eye-View (BEV) representation has emerged as a crucial aspect of 3D object detection in autonomous driving. However, existing methods are susceptible to the inaccurate calibration relationship between LiDAR and the camera sensor. Such inaccuracies result in errors in depth estimation for the camera branch, ultimately causing misalignment between LiDAR and camera BEV features. In this work, we propose a robust fusion framework called Graph BEV. Addressing errors caused by inaccurate point cloud projection, we introduce a Local Align module that employs neighbor-aware depth features via Graph matching. Additionally, we propose a Global Align module to rectify the misalignment between LiDAR and camera BEV features. Our Graph BEV framework achieves state-of-the-art performance, with an mAP of 70.1\%, surpassing BEV Fusion by 1.6\% on the nuscenes validation set. Importantly, our Graph BEV outperforms BEV Fusion by 8.3\% under conditions with misalignment noise.

The class-conditional image generation based on diffusion models is renowned for generating high-quality and diverse images. However, most prior efforts focus on generating images for general categories, e.g., 1000 classes in ImageNet-1k. A more challenging task, large-scale fine-grained image generation, remains the boundary to explore. In this work, we present a parameter-efficient strategy, called FineDiffusion, to fine-tune large pre-trained diffusion models scaling to large-scale fine-grained image generation with 10,000 categories. FineDiffusion significantly accelerates training and reduces storage overhead by only fine-tuning tiered class embedder, bias terms, and normalization layers' parameters. To further improve the image generation quality of fine-grained categories, we propose a novel sampling method for fine-grained image generation, which utilizes superclass-conditioned guidance, specifically tailored for fine-grained categories, to replace the conventional classifier-free guidance sampling. Compared to full fine-tuning, FineDiffusion achieves a remarkable 1.56x training speed-up and requires storing merely 1.77% of the total model parameters, while achieving state-of-the-art FID of 9.776 on image generation of 10,000 classes. Extensive qualitative and quantitative experiments demonstrate the superiority of our method compared to other parameter-efficient fine-tuning methods. The code and more generated results are available at our project website: https://finediffusion.github.io/.

In the realm of modern autonomous driving, the perception system is indispensable for accurately assessing the state of the surrounding environment, thereby enabling informed prediction and planning. Key to this system is 3D object detection methods, that utilize vehicle-mounted sensors such as LiDAR and cameras to identify the size, category, and location of nearby objects. Despite the surge in 3D object detection methods aimed at enhancing detection precision and efficiency, there is a gap in the literature that systematically examines their resilience against environmental variations, noise, and weather changes. This study emphasizes the importance of robustness, alongside accuracy and latency, in evaluating perception systems under practical scenarios. Our work presents an extensive survey of camera-based, LiDAR-based, and multimodal 3D object detection algorithms, thoroughly evaluating their trade-off between accuracy, latency, and robustness, particularly on datasets like KITTI-C and nuScenes-C to ensure fair comparisons. Among these,multimodal 3D detection approaches exhibit superior robustness and a novel taxonomy is introduced to reorganize its literature for enhanced clarity. This survey aims to offer a more practical perspective on the current capabilities and constraints of 3D object detection algorithms in real-world applications, thus steering future research towards robustness-centric advancements

LiDAR-camera fusion can enhance the performance of 3D object detection by utilizing complementary information between depth-aware LiDAR points and semantically rich images. Existing voxel-based methods face significant challenges when fusing sparse voxel features with dense image features in a one-to-one manner, resulting in the loss of the advantages of images, including semantic and continuity information, leading to sub-optimal detection performance, especially at long distances. In this paper, we present VoxelNextFusion, a multi-modal 3D object detection framework specifically designed for voxel-based methods, which effectively bridges the gap between sparse point clouds and dense images. In particular, we propose a voxel-based image pipeline that involves projecting point clouds onto images to obtain both pixel- and patch-level features. These features are then fused using a self-attention to obtain a combined representation. Moreover, to address the issue of background features present in patches, we propose a feature importance module that effectively distinguishes between foreground and background features, thus minimizing the impact of the background features. Extensive experiments were conducted on the widely used KITTI and nuScenes 3D object detection benchmarks. Notably, our VoxelNextFusion achieved around +3.20% in AP@0.7 improvement for car detection in hard level compared to the Voxel R-CNN baseline on the KITTI test dataset

With the development of social media, rumors have been spread broadly on social media platforms, causing great harm to society. Beside textual information, many rumors also use manipulated images or conceal textual information within images to deceive people and avoid being detected, making multimodal rumor detection be a critical problem. The majority of multimodal rumor detection methods mainly concentrate on extracting features of source claims and their corresponding images, while ignoring the comments of rumors and their propagation structures. These comments and structures imply the wisdom of crowds and are proved to be crucial to debunk rumors. Moreover, these methods usually only extract visual features in a basic manner, seldom consider tampering or textual information in images. Therefore, in this study, we propose a novel Vision and Graph Fused Attention Network (VGA) for rumor detection to utilize propagation structures among posts so as to obtain the crowd opinions and further explore visual tampering features, as well as the textual information hidden in images. We conduct extensive experiments on three datasets, demonstrating that VGA can effectively detect multimodal rumors and outperform state-of-the-art methods significantly.

Multi-modal 3D object detectors are dedicated to exploring secure and reliable perception systems for autonomous driving (AD).Although achieving state-of-the-art (SOTA) performance on clean benchmark datasets, they tend to overlook the complexity and harsh conditions of real-world environments. With the emergence of visual foundation models (VFMs), opportunities and challenges are presented for improving the robustness and generalization of multi-modal 3D object detection in AD. Therefore, we propose RoboFusion, a robust framework that leverages VFMs like SAM to tackle out-of-distribution (OOD) noise scenarios. We first adapt the original SAM for AD scenarios named SAM-AD. To align SAM or SAM-AD with multi-modal methods, we then introduce AD-FPN for upsampling the image features extracted by SAM. We employ wavelet decomposition to denoise the depth-guided images for further noise reduction and weather interference. At last, we employ self-attention mechanisms to adaptively reweight the fused features, enhancing informative features while suppressing excess noise. In summary, RoboFusion significantly reduces noise by leveraging the generalization and robustness of VFMs, thereby enhancing the resilience of multi-modal 3D object detection. Consequently, RoboFusion achieves SOTA performance in noisy scenarios, as demonstrated by the KITTI-C and nuScenes-C benchmarks. Code is available at https://github.com/adept-thu/RoboFusion.

In the field of 3D object detection tasks, fusing heterogeneous features from LiDAR and camera sensors into a unified Bird's Eye View (BEV) representation is a widely adopted paradigm. However, existing methods are often compromised by imprecise sensor calibration, resulting in feature misalignment in LiDAR-camera BEV fusion. Moreover, such inaccuracies result in errors in depth estimation for the camera branch, ultimately causing misalignment between LiDAR and camera BEV features. In this work, we propose a novel ContrastAlign approach that utilizes contrastive learning to enhance the alignment of heterogeneous modalities, thereby improving the robustness of the fusion process. Specifically, our approach includes the L-Instance module, which directly outputs LiDAR instance features within LiDAR BEV features. Then, we introduce the C-Instance module, which predicts camera instance features through RoI (Region of Interest) pooling on the camera BEV features. We propose the InstanceFusion module, which utilizes contrastive learning to generate similar instance features across heterogeneous modalities. We then use graph matching to calculate the similarity between the neighboring camera instance features and the similarity instance features to complete the alignment of instance features. Our method achieves state-of-the-art performance, with an mAP of 70.3%, surpassing BEVFusion by 1.8% on the nuScenes validation set. Importantly, our method outperforms BEVFusion by 7.3% under conditions with misalignment noise.

The Grey Wolf Optimizer (GWO) is recognized as a novel meta-heuristic algorithm inspired by the social leadership hierarchy and hunting mechanism of grey wolves. It is well-known for its simple parameter setting, fast convergence speed, and strong optimization capability. In the original GWO, there are two significant design flaws in its fundamental optimization mechanisms. Problem (1): the algorithm fails to inherit from elite positions from the last iteration when generating the next positions of the wolf population, potentially leading to suboptimal solutions. Problem (2): the positions of the population are updated based on the central position of the three leading wolves (alpha, beta, delta), without a balanced mechanism between local and global search. To tackle these problems, an enhanced Grey Wolf Optimizer with Elite Inheritance Mechanism and Balance Search Mechanism, named as EBGWO, is proposed to improve the effectiveness of the position updating and the quality of the convergence solutions. The IEEE CEC 2014 benchmark functions suite and a series of simulation tests are employed to evaluate the performance of the proposed algorithm. The simulation tests involve a comparative study between EBGWO, three GWO variants, GWO and two well-known meta-heuristic algorithms. The experimental results demonstrate that the proposed EBGWO algorithm outperforms other meta-heuristic algorithms in both accuracy and convergence speed. Three engineering optimization problems are adopted to prove its capability in processing real-world problems. The results indicate that the proposed EBGWO outperforms several popular algorithms., Comment: 51 pages, 21 tables, 16 figures, journal

The integration of Foundation Models (FMs) with Federated Learning (FL) presents a transformative paradigm in Artificial Intelligence (AI), offering enhanced capabilities while addressing concerns of privacy, data decentralization, and computational efficiency. This paper provides a comprehensive survey of the emerging field of Federated Foundation Models (FedFM), elucidating their synergistic relationship and exploring novel methodologies, challenges, and future directions that the FL research field needs to focus on in order to thrive in the age of foundation models. A systematic multi-tiered taxonomy is proposed, categorizing existing FedFM approaches for model training, aggregation, trustworthiness, and incentivization. Key challenges, including how to enable FL to deal with high complexity of computational demands, privacy considerations, contribution evaluation, and communication efficiency, are thoroughly discussed. Moreover, the paper explores the intricate challenges of communication, scalability and security inherent in training/fine-tuning FMs via FL, highlighting the potential of quantum computing to revolutionize the training, inference, optimization and data encryption processes. This survey underscores the importance of further research to propel innovation in FedFM, emphasizing the need for developing trustworthy solutions. It serves as a foundational guide for researchers and practitioners interested in contributing to this interdisciplinary and rapidly advancing field., Comment: Survey of Federated Foundation Models (FedFM)

Optical coherence tomography (OCT) is an optical imaging method based on low-coherence interferometry. OCT uses light waves to take cross-section images of your internal microstructure such as retina by measuring light backscattered from the sample. It can provide reliable and high resolution images of biological tissues at different levels. Real-time OCT can be a challenge due to the heavy computational load required to process acquired data streams. Reconstructing functional images with multi-functional OCT imaging requires additional processing, further increasing processing time. So we introduced graphics processing unit (GPU) processing and some code opti- mization strategies of MATLAB to accelerate the image processing. Also, we implemented a web application to run MATLAB program remotely for the convenience of the researchers. It can allow multiple users to remote control MATLAB programs and do some basic oper- ations through both mobile phone and computer. In Chapter1, I will explain the OCT data processing workflow and describe the motivation of OCT data processing acceleration. vi Chapter2 will mainly focus on the methods that I used to improve the performance of the MATLAB code and show the comparison of the total processing time. In Chapter3, I will introduce the features of the web application as well as the structure of the whole application for both front-end and back-end. Chapter4 is the conclusion part of the entire post-processing acceleration project, some future work will also be listed in chapter4.

Background: Despite an increase in popularity, the meal kit delivery service is facing the challenge of high cancellations and difficulty with customer retention. As it is a comparatively novel concept in the context of digital food retail, more scholarly research should be conducted. Existing research on meal kit delivery services is limited, focusing on customer-perceived benefits and product attributes, such as food quality, menu variety, health orientation, convenience, and price, associated with purchasing intention. Therefore, it is an insightful study to understand the factors that influence customers’ brand loyalty in relation to meal kit delivery services. Purpose: The purpose of this study aims to address the research gap in understanding customer behaviours associated with brand loyalty in the meal kit delivery services industry. From an academic perspective, it seeks to understand the factors influencing customer behaviours towards brand loyalty in the context of digitalisation in food retail. Additionally, the study aims to develop a conceptual framework that clarifies the factors consumers consider when demonstrating brand loyalty to meal kit delivery services. In terms of practical contribution, the findings aim to provide valuable insights to meal kit delivery companies, assisting them in effectively addressing the challenge of customer retention. Method: This is an exploratory qualitative case study based on HelloFresh that includes 13 in-depth semi-structured interviews to collect rich and insightful data. The qualitative approach of interviews i created primary data for empirical findings which were further coded, analysed, and identified inductively using thematic analysis. Conclusion: The study contributes to the existing literature by examining customer behaviours and factors influencing brand loyalty in meal kit delivery services. The findings indicated that both behavioural and attitudinal elements play a significant role in custo

Human visual grouping processes consolidate independent visual objects into grouped visual features on the basis of shared characteristics; these visual features can themselves be grouped, resulting in a hierarchical representation of visual grouping information. In the present study, participants provided free-response groupings of a set of stimuli that contained consistent structural relationships between a limited set of visual features. These grouping patterns were evaluated for relationships between specific visual features and the participant’s grouping patterns. We observed that the relative size of the visual features differentiated groupings across levels of the grouping hierarchy, while the form of visual objects and features distinguished separate groups within a particular level of hierarchy. These consistent relationships between visual feature characteristics and placement within a grouping hierarchy can be leveraged to advance computational theories of human visual grouping behavior, which can in turn be applied to effective design for interfaces such as voter ballots.

Keyan Hu,1,2,* Xueying Liu,1,* Hongfeng Chang,1 Yi Zhang,1 Hui Zhou,3 Lei Liu,1 Xin Zhang,1 Ziying Jiao,4 Bing Shen,4 Qiu Zhang1 1Department of Endocrinology, The First Affiliated Hospital of Anhui Medical University, Hefei, Anhui, 230022, People’s Republic of China; 2Department of Endocrinology, The First Affiliated Hospital of Henan University of Science and Technology, Luoyang, Henan, 471003, People’s Republic of China; 3Institute of Brain Diseases and Cognition, School of Medicine, Xiamen University, Xiamen, Fujian, 361000, People’s Republic of China; 4School of Basic Medical Sciences, Anhui Medical University, Hefei, Anhui, 230032, People’s Republic of China*These authors contributed equally to this workCorrespondence: Qiu Zhang, Department of Endocrinology, The First Affiliated Hospital of Anhui Medical University, 218 Jixi Road, Hefei, Anhui, 230022, People’s Republic of China, Tel/Fax +86-551-62923631, Email zhangqiu@ahmu.edu.cn Bing Shen, Department of Physiology, School of Basic Medical Sciences, Anhui Medical University, 81 Meishan Road, Hefei, Anhui, 230032, People’s Republic of China, Tel +86-551-65161132, Fax +86-551-65161126, Email shenbing@ahmu.edu.cnIntroduction: Diabetes foot ulcer (DFU) is a serious complication of diabetes characterized with chronic foot ulceration, poor wound healing (WH), and persistent inflammation. MiR-221-3p, as microRNA, has been shown to accelerate WH in previous study, but the underlying mechanisms are poorly understood.Methods: In this study, we aimed to determine how miR-221-3p influences WH by targeting THBS1. The effect of miRNA-221-3p on wound healing of diabetes by epidermal injection of miRNA-221-3p agomir. In vitro generated human immortalized keratinocytes (HaCaT cells) were transfected with miR-mimics and negative control with high glucose treatment. The effects of miRNA-221-3p on cell apoptosis and angiogenesis using cell apoptosis assay and the tube formation assay, respecti

Background With advances in massive parallel sequencing (MPS) technology, whole-genome sequencing (WGS) has gradually evolved into the first-tier diagnostic test for genetic disorders. However, deployment practice and pipeline testing for clinical WGS are lacking. Methods In this study, we introduced a whole WGS pipeline for genetic disorders, which included the entire process from obtaining a sample to clinical reporting. All samples that underwent WGS were constructed using polymerase chain reaction (PCR)-free library preparation protocols and sequenced on the MGISEQ-2000 platform. Bioinformatics pipelines were developed for the simultaneous detection of various types of variants, including single nucleotide variants (SNVs), insertions and deletions (indels), copy number variants (CNVs) and balanced rearrangements, mitochondrial (MT) variants, and other complex variants such as repeat expansion, pseudogenes and absence of heterozygosity (AOH). A semiautomatic pipeline was developed for the interpretation of potential SNVs and CNVs. Forty-five samples (including 14 positive commercially available samples, 23 laboratory-held positive cell lines and 8 clinical cases) with known variants were used to validate the whole pipeline. Results In this study, a whole WGS pipeline for genetic disorders was developed and optimized. Forty-five samples with known variants (6 with SNVs and Indels, 3 with MT variants, 5 with aneuploidies, 1 with triploidy, 23 with CNVs, 5 with balanced rearrangements, 2 with repeat expansions, 1 with AOHs, and 1 with exon 7-8 deletion of SMN1 gene) validated the effectiveness of our pipeline. Conclusions This study has been piloted in test development, optimization, and validation of the WGS pipeline for genetic disorders. A set of best practices were recommended using our pipeline, along with a dataset of positive samples for benchmarking.

Recently, three optical tidal disruption event (TDE) candidates discovered by the Zwicky Transient Facility (ZTF) have been suggested to be coincident with high-energy neutrinos. They all exhibit unusually strong dust infrared (IR) echoes, with their peak times matching the neutrino arrival time even better than the optical peaks. We hereby report on two new TDE candidates that are spatially and temporally coincident with neutrinos by matching our sample of mid-infrared outbursts in nearby galaxies (MIRONG) with Gold alerts of IceCube high-energy neutrino events up to June 2022. The two candidates show negligible optical variability according to their ZTF light curves and can therefore be classified as part of the growing population of obscured TDE candidates. The chance probability of finding two such candidates about $\sim3\%$ by redistributing the MIRONG sources randomly in the SDSS footprint, which will be as low as $\sim0.1\%$ (or $\sim0.2\%$) if we limit to sources with increased fluxes (or variability amplitudes) comparable with the matched two sources. Our findings further support the potential connection between high-energy neutrinos and TDEs in dusty environments by increasing the total number of neutrino-associated TDE and TDE candidates to five, although the underlying physics remains poorly understood., Comment: Published, ApJL, 953, L12

Blind face restoration usually synthesizes degraded low-quality data with a pre-defined degradation model for training, while more complex cases could happen in the real world. This gap between the assumed and actual degradation hurts the restoration performance where artifacts are often observed in the output. However, it is expensive and infeasible to include every type of degradation to cover real-world cases in the training data. To tackle this robustness issue, we propose Diffusion-based Robust Degradation Remover (DR2) to first transform the degraded image to a coarse but degradation-invariant prediction, then employ an enhancement module to restore the coarse prediction to a high-quality image. By leveraging a well-performing denoising diffusion probabilistic model, our DR2 diffuses input images to a noisy status where various types of degradation give way to Gaussian noise, and then captures semantic information through iterative denoising steps. As a result, DR2 is robust against common degradation (e.g. blur, resize, noise and compression) and compatible with different designs of enhancement modules. Experiments in various settings show that our framework outperforms state-of-the-art methods on heavily degraded synthetic and real-world datasets., Comment: Accepted to CVPR 2023

Juan Yang,1 Kyung-Min Shin,2 Alex Do,1 Dennis M Bierle,1 Abd Moain Abu Dabrh,3 Ziying Yin,4 Brent A Bauer,1 Arya B Mohabbat1 1Division of General Internal Medicine, Mayo Clinic, Rochester, MN, 55905, USA; 2KM Science Research Division, Korea Institute of Oriental Medicine, Daejeon, 34054, South Korea; 3Division of General Internal Medicine, Mayo Clinic, Jacksonville, FL, 32224, USA; 4Department of Radiology, Mayo Clinic, Rochester, MN, 55905, USACorrespondence: Arya B Mohabbat, Division of General Internal Medicine, Mayo Clinic, 200 First Street SW, RO_GO_17_102GIM, Rochester, MN, 55905, USA, Tel +1-507-284-2511, Email Mohabbat.Arya@mayo.eduAbstract: Fibromyalgia (FM) is a chronic pain sensitivity syndrome characterized by diffuse musculoskeletal pain and many other systemic manifestations. Low-dose naltrexone (LDN) has been increasingly used as an off-label treatment option in FM. However, current evidence on the safety and efficacy of LDN in patients with FM is not well known. To systematically assess the current evidence on the safety and efficacy of LDN use in the treatment of FM. A comprehensive bibliographic search was conducted on EBM Reviews – Cochrane Central Register of Controlled Trials, EBM Reviews – Cochrane Database of Systematic, Embase, Ovid MEDLINE(R) and Epub Ahead of Print, In-Process, In-Data-Review & Other Non-Indexed Citations, Daily and Versions and Scopus databases in September 2022. Inclusion criteria were articles that were published in English, focusing on clinical trials involving LDN for the treatment of FM. Two reviewers independently screened and extracted the data. A qualitative analysis was used due to the high methodological heterogeneity between studies. The electronic search produced 805 articles. After applying the inclusion criteria, 9 articles (one RCT, two case reports, two case series, and four pilot trials) were selected for evaluation. LDN intervention protocols, study designs, and follow-up periods were different a

Non-maximum suppression (NMS) is an essential post-processing module used in many 3D object detection frameworks to remove overlapping candidate bounding boxes. However, an overreliance on classification scores and difficulties in determining appropriate thresholds can affect the resulting accuracy directly. To address these issues, we introduce fuzzy learning into NMS and propose a novel generalized Fuzzy-NMS module to achieve finer candidate bounding box filtering. The proposed Fuzzy-NMS module combines the volume and clustering density of candidate bounding boxes, refining them with a fuzzy classification method and optimizing the appropriate suppression thresholds to reduce uncertainty in the NMS process. Adequate validation experiments are conducted using the mainstream KITTI and large-scale Waymo 3D object detection benchmarks. The results of these tests demonstrate the proposed Fuzzy-NMS module can improve the accuracy of numerous recently NMS-based detectors significantly, including PointPillars, PV-RCNN, and IA-SSD, etc. This effect is particularly evident for small objects such as pedestrians and bicycles. As a plug-and-play module, Fuzzy-NMS does not need to be retrained and produces no obvious increases in inference time.

Radar has stronger adaptability in adverse scenarios for autonomous driving environmental perception compared to widely adopted cameras and LiDARs. Compared with commonly used 3D radars, the latest 4D radars have precise vertical resolution and higher point cloud density, making it a highly promising sensor for autonomous driving in complex environmental perception. However, due to the much higher noise than LiDAR, manufacturers choose different filtering strategies, resulting in an inverse ratio between noise level and point cloud density. There is still a lack of comparative analysis on which method is beneficial for deep learning-based perception algorithms in autonomous driving. One of the main reasons is that current datasets only adopt one type of 4D radar, making it difficult to compare different 4D radars in the same scene. Therefore, in this paper, we introduce a novel large-scale multi-modal dataset featuring, for the first time, two types of 4D radars captured simultaneously. This dataset enables further research into effective 4D radar perception algorithms.Our dataset consists of 151 consecutive series, most of which last 20 seconds and contain 10,007 meticulously synchronized and annotated frames. Moreover, our dataset captures a variety of challenging driving scenarios, including many road conditions, weather conditions, nighttime and daytime with different lighting intensities and periods. Our dataset annotates consecutive frames, which can be applied to 3D object detection and tracking, and also supports the study of multi-modal tasks. We experimentally validate our dataset, providing valuable results for studying different types of 4D radars. This dataset is released on https://github.com/adept-thu/Dual-Radar.

LiDAR and cameras are complementary sensors for 3D object detection in autonomous driving. However, it is challenging to explore the unnatural interaction between point clouds and images, and the critical factor is how to conduct feature alignment of heterogeneous modalities. Currently, many methods achieve feature alignment by projection calibration only, without considering the problem of coordinate conversion accuracy errors between sensors, leading to sub-optimal performance. In this paper, we present GraphAlign, a more accurate feature alignment strategy for 3D object detection by graph matching. Specifically, we fuse image features from a semantic segmentation encoder in the image branch and point cloud features from a 3D Sparse CNN in the LiDAR branch. To save computation, we construct the nearest neighbor relationship by calculating Euclidean distance within the subspaces that are divided into the point cloud features. Through the projection calibration between the image and point cloud, we project the nearest neighbors of point cloud features onto the image features. Then by matching the nearest neighbors with a single point cloud to multiple images, we search for a more appropriate feature alignment. In addition, we provide a self-attention module to enhance the weights of significant relations to fine-tune the feature alignment between heterogeneous modalities. Extensive experiments on nuScenes benchmark demonstrate the effectiveness and efficiency of our GraphAlign.

For learning-based sound event localization and detection (SELD) methods, different acoustic environments in the training and test sets may result in large performance differences in the validation and evaluation stages. Different environments, such as different sizes of rooms, different reverberation times, and different background noise, may be reasons for a learning-based system to fail. On the other hand, acquiring annotated spatial sound event samples, which include onset and offset time stamps, class types of sound events, and direction-of-arrival (DOA) of sound sources is very expensive. In addition, deploying a SELD system in a new environment often poses challenges due to time-consuming training and fine-tuning processes. To address these issues, we propose Meta-SELD, which applies meta-learning methods to achieve fast adaptation to new environments. More specifically, based on Model Agnostic Meta-Learning (MAML), the proposed Meta-SELD aims to find good meta-initialized parameters to adapt to new environments with only a small number of samples and parameter updating iterations. We can then quickly adapt the meta-trained SELD model to unseen environments. Our experiments compare fine-tuning methods from pre-trained SELD models with our Meta-SELD on the Sony-TAU Realistic Spatial Soundscapes 2023 (STARSSS23) dataset. The evaluation results demonstrate the effectiveness of Meta-SELD when adapting to new environments., Comment: Submitted to DCASE 2023 Workshop

Plasmonic hot electrons are electrons with high kinetic energy, generated from theplasmonic nanostructures. The application of hot electrons has been widely studied in thecommunity of photochemistry and optoelectronics. Many applications like photoelectrochemistryand photodetector involve semiconductors, and these applications are plagued by low hotelectron injection efficiency in the metal-semiconductor junctions which hinders the widerapplications for the hot electrons.Since the exfoliation of graphene with scotch tape in 2004, two dimensional (2D) materialshave been widely studied for their unique properties when the thickness scales down to atomicallythin. Transition metal dichalcogenides are a class 2D materials, they are semiconductors, andthey have the different band structures with different material compositions. For each type of thetransition metal dichalcogenide, its few-layer counterparts have both direct band transition andthe indirect band transition, this unique band structure of the few-layer 2D transition metaldichalcogenides opens up the possibilities for studying the relationship between the hot electroninjection and the band structure in the metal-semiconductor junction.Inspired by the unique band structure of the 2D semiconductors, we design the structureformed with plasmonic nanostructures and 2D semiconductors as a model system to exploreplasmonic hot electron injection process at the metal-semiconductor junction, in which we employhigh mobility 2D semiconductor to capture the hot electrons. Due to the high photoluminescencequantum yield WSe2, photoluminescence spectra are sued to probe the hot electron injectionmechanism between the gold and few-layer WSe2. We demonstrated that the hot electrons tendto at first inject into the energy lower L point, and then to the K point of the of the few-layer WSe2.Another question considered in this dissertation is how to modulate the hot electroninjection to improve hot electron collection in the semicond

In parallel with the rapid adoption of artificial intelligence (AI) empowered by advances in AI research, there has been growing awareness and concerns of data privacy. Recent significant developments in the data regulation landscape have prompted a seismic shift in interest toward privacy-preserving AI. This has contributed to the popularity of Federated Learning (FL), the leading paradigm for the training of machine learning models on data silos in a privacy-preserving manner. In this survey, we explore the domain of personalized FL (PFL) to address the fundamental challenges of FL on heterogeneous data, a universal characteristic inherent in all real-world datasets. We analyze the key motivations for PFL and present a unique taxonomy of PFL techniques categorized according to the key challenges and personalization strategies in PFL. We highlight their key ideas, challenges, opportunities, and envision promising future trajectories of research toward a new PFL architectural design, realistic PFL benchmarking, and trustworthy PFL approaches. IEEE

Tian X, Zeng A, Liu Z, et al. Int J Nanomedicine. 2020;15:6519–6529. The authors have advised that there was an error in the Funding section on page 6528. Line 4, the text “(grant Q195026028)” should read “(grant 2019A1515011800)”. The authors apologize for this error. Read the original article

Background Elevated red cell distribution width (RDW) has been associated with degenerative conditions in aging. Aims We aimed to evaluate the associations of RDW and anemia with lower-extremity physical function among rural-dwelling older adults. Methods This population-based cross-sectional study included 5093 rural residents (age >= 60 years, 57.3% women) who participated in the MIND-China Study in Shandong. Data were collected via face-to-face interviews, clinical examinations, and laboratory tests. RDW was categorized according to quartiles and the lower-extremity physical function was assessed using the Short Physical Performance Battery (SPPB), Results Multiple linear regression analyses suggested that the fourth quartile of RDW (vs. first quartile) was associated with lower SPPB summary score (beta-coefficient - 0.38; 95% CI - 0.58 to - 0.18) and lower scores in balance test (- 0.09; - 0.17 to - 0.01), chair stand test (- 0.17; - 0.27 to - 0.07), and walking speed test (- 0.12; - 0.19 to - 0.05). Anemia was associated with a multiple-adjusted beta-coefficient of - 0.34 (- 0.52 to - 0.16) for SPPB summary score. Stratified analysis by anemia showed that there was a linear association between RDW and SPPB in individuals without anemia but a J-shaped association in individuals with anemia. Discussion This large-scale population-based study revealed the associations of high RDW and anemia with poor lower-extremity physical function among rural-dwelling Chinese older adults. These findings suggest that an elevated RDW might be a biochemical marker for poor lower-extremity physical function among older adults. Conclusions Anemia and an elevated RDW are associated with poor performance in lower-extremity physical function among rural-dwelling Chinese older adults.

Sedentary behaviour is associated with a range of adverse health conditions. Population-based studies have rarely examined the distribution and associated factors of accelerometer-measured sedentary behaviour patterns in rural-dwelling older adults. This population-based study included 2096 rural-dwelling older adults (age ≥60 years; 59.0% women) derived from baseline participants of the MIND-China Study. Total sedentary time and patterns (e.g., uninterrupted bouts and breaks) were derived from the hip-worn accelerometers for 7 days. Physical function was assessed using the Short Physical Performance Battery test. Data were analysed using general linear models. Overall, participants spent 58.8% of daily waking time in sedentary behaviour, with nearly half of sedentary time being accumulated through sedentary bouts of 30+ minutes. Men spent more total and accumulated sedentary time than women in each sedentary bout duration, while women had more daily 1+ minute sedentary bouts than men (all P < 0.001). Controlling for moderate-to-vigorous physical activity and other confounders, more prolonged sedentary time and fewer breaks were significantly associated with poor physical function, balance, lower limb strength, and walking speed (all P < 0.001). In older adults living in rural communities, prolonged sedentary behaviour and less frequent breaks are associated with poor physical function.

Purpose: We explore the associations of individual and composite cardiovascular health metrics with all-cause dementia, Alzheimer’s disease, and vascular dementia among rural-dwelling older adults and the potential age variations in their associations. Patients and Methods: This community-based cross-sectional study included 4980 older adults (age ≥ 65 years; 57.23% women) from the baseline examination of MIND-China. In March–September 2018, data were collected via face-to-face interviews, clinical examinations, and laboratory test. We defined six cardiovascular health metrics according to the modified American Heart Association’s recommendations. We diagnosed dementia and its subtypes following the international criteria. Data were analyzed using logistic regression models. Results: Of all the participants, 250 were diagnosed with dementia, including 165 with Alzheimer’s disease and 75 with vascular dementia. Ideal composite global cardiovascular health metrics (vs poor composite metrics) were associated with a multi-adjusted odds ratio (95% confidence interval) of 0.62 (0.42– 0.93) for dementia, 0.88 (0.52– 1.48) for Alzheimer’s disease, and 0.31 (0.16– 0.60) for vascular dementia. Moreover, ideal biological cardiovascular health metrics were associated with multi-adjusted odds ratio of 0.52 (0.28– 0.95) for dementia and 0.21 (0.06– 0.77) for vascular dementia in young–old adults (65– 74 years), whereas ideal behavioral cardiovascular health metrics were associated with multi-adjusted odds ratio of 0.48 (0.26– 0.89) for dementia and 0.16 (0.06– 0.43) for vascular dementia in old–old adults (≥ 75 years). Conclusion: Our results suggest that ideal cardiovascular health metrics are cross-sectionally associated with a low likelihood of dementia and vascular dementia among rural-dwelling older Chinese adults. The associations vary with age, components of cardiovascular health metrics, and dementia subtypes.

Contains fulltext : 253267.pdf (Publisher’s version ) (Open Access)

The sodium super-ionic conductor (NASICON)-type solid-state electrolyte Li1.4Al0.4Ti1.6(PO4)3 (LATP) is an attractive alternative to liquid electrolytes for lithium batteries. The rapid development of LATP, however, is hindered by its poor interfacial compatibilities against the Li metal. Herein, a flexible membrane coating layer consisting of Mg3N2 and PVDF has been adopted to modify LATP via a simple drop-casting method. A multifunctional interlayer with Mg, LiF and Li3N is in situ constructed by the reaction of the coating layer with the Li metal. The decomposition of LATP has been restrained and interfacial ionic transport kinetics has been improved with the modification. Benefitting from the multifunctional interlayer, the critical current density of LATP is improved from 0.34 mA cm−2 to 0.76 mA cm−2. The symmetric cells assembled with the modified LATP exhibit a stable cycle for more than 1000 h at 0.20 mA cm−2, and the Li/LiFePO4 cells after modification have a capacity retention of 80% after 385 cycles at 2C. The present work demonstrates a promising strategy for fine interfacial stability tuning and low-impedance LATP., QC 20230704

Guichan Liao, Ziying Liu, Muye Xia, Hongjie Chen, Houji Wu, Bing Li, Tao Yu, Shaohang Cai, Xiaoyong Zhang, Jie Peng State Key Laboratory of Organ Failure Research, Guangdong Provincial Key Laboratory of Viral Hepatitis Research, Department of Infectious Diseases, Nanfang Hospital, Southern Medical University, Guangzhou, People’s Republic of ChinaCorrespondence: Jie Peng, State Key Laboratory of Organ Failure Research, Guangdong Provincial Key Laboratory of Viral Hepatitis Research, Department of Infectious Diseases, Nanfang Hospital, Southern Medical University, Guangzhou, People’s Republic of China, Tel +86 20 6278 7428, Fax +86 20 8771 9653, Email pjie138@163.comPurpose: The immunoinhibitory receptor, programmed death 1 (PD-1), plays a critical role in immune suppression during chronic viral infection. The significance of circulating soluble PD-1 (sPD-1) in patients with chronic hepatitis B who have discontinued long-term nucleos(t)ide analog (NA) treatment remains unknown.Patients and Methods: A prospective cohort study was conducted using serial blood samples from chronic hepatitis B patients who discontinued long-term NA treatment. The current analysis included 115 non-cirrhotic patients with HBV DNA negative and HBsAg positive at the moment of NA discontinuation. Levels of sPD-1 were measured in all available samples using sandwich enzyme-linked immunoassay.Results: Sixty-two patients experienced a clinical relapse and 14 occurred HBsAg loss, with 8-year cumulative rates of 56.6% and 23.4%, respectively. Time-dependent receiver operating characteristic curve analysis for sPD-1 derived 156 pg/mL, which is equivalent to the detectable threshold, as an optimal cut-off value for predicting 8-year clinical relapse. Patients with detectable sPD-1 at end of treatment (EOT) had a significant lower incidence of clinical relapse (48% vs 67%, hazard ratio [HR] 0.454, p = 0.006), but a remarkable higher probability of HBsAg loss (33.7% vs 2.4%, HR 9.17, p = 0.038), co

To enhance children's health, the promotion of nutrition literacy in school is vital as it helps prevent the development of health conditions and diseases and maintain healthy lifestyles. Taiwan features the top highest ratio of convenience stores per population density. Convenience stores, an increasingly popular dining place, were linked to the development of eating behavior and body weight issues in children. An eight-week classroom-based nutrition intervention, employing the Traffic Light Diet as a framework, targeting children's perception of and intention to visit the convenience store was implemented. The study conducted a quasi-experimental pretest-posttest research design with a comparison group. A total of 49 students participated in the study, with 25 in the intervention and 24 in the comparison group. Data were collected by utilizing surveys, interviews, and observations. The study's findings demonstrated the positive trajectory of the impact of this intervention on increasing food-and-nutrition-related knowledge and improving healthier diet choices at convenience stores among children. One main theme was identified in coding interviews: parent involvement in meal preparation may reduce convenience store use and increase consumption of vegetables among children. Assessing the influence of parental support for healthy dietary choices, eating nutritious foods at home, and involving the family in meal preparation is an area for future research.

LiDAR and camera fusion techniques are promising for achieving 3D object detection in autonomous driving. Most multi-modal 3D object detection frameworks integrate semantic knowledge from 2D images into 3D LiDAR point clouds to enhance detection accuracy. Nevertheless, the restricted resolution of 2D feature maps impedes accurate re-projection and often induces a pronounced boundary-blurring effect, which is primarily attributed to erroneous semantic segmentation. To well handle this limitation, we propose a general multi-modal fusion framework Multi-Sem Fusion (MSF) to fuse the semantic information from both the 2D image and 3D points scene parsing results. Specifically, we employ 2D/3D semantic segmentation methods to generate the parsing results for 2D images and 3D point clouds. The 2D semantic information is further reprojected into the 3D point clouds with calibration parameters. To handle the misalignment between the 2D and 3D parsing results, we propose an Adaptive Attention-based Fusion (AAF) module to fuse them by learning an adaptive fusion score. Then the point cloud with the fused semantic label is sent to the following 3D object detectors. Furthermore, we propose a Deep Feature Fusion (DFF) module to aggregate deep features at different levels to boost the final detection performance. The effectiveness of the framework has been verified on two public large-scale 3D object detection benchmarks by comparing them with different baselines. The experimental results show that the proposed fusion strategies can significantly improve the detection performance compared to the methods using only point clouds and the methods using only 2D semantic information. Most importantly, the proposed approach significantly outperforms other approaches and sets state-of-the-art results on the nuScenes testing benchmark.

Real-world image denoising is a practical image restoration problem that aims to obtain clean images from in-the-wild noisy inputs. Recently, the Vision Transformer (ViT) has exhibited a strong ability to capture long-range dependencies, and many researchers have attempted to apply the ViT to image denoising tasks. However, a real-world image is an isolated frame that makes the ViT build long-range dependencies based on the internal patches, which divides images into patches, disarranges noise patterns and damages gradient continuity. In this article, we propose to resolve this issue by using a continuous Wavelet Sliding-Transformer that builds frequency correspondences under real-world scenes, called DnSwin. Specifically, we first extract the bottom features from noisy input images by using a convolutional neural network (CNN) encoder. The key to DnSwin is to extract high-frequency and low-frequency information from the observed features and build frequency dependencies. To this end, we propose a Wavelet Sliding-Window Transformer (WSWT) that utilizes the discrete wavelet transform (DWT), self-attention and the inverse DWT (IDWT) to extract deep features. Finally, we reconstruct the deep features into denoised images using a CNN decoder. Both quantitative and qualitative evaluations conducted on real-world denoising benchmarks demonstrate that the proposed DnSwin performs favorably against the state-of-the-art methods., Comment: Accepted by KBS; Wavelet downsampling expands window size in Transformer cheaply for a better real-world denosing

In EUROCRYPT 2018, Cid $et\;al.$ introduced a new concept on the cryptographic property of S-boxes to evaluate the subtleties of boomerang-style attacks. This concept was named as boomerang connectivity table (BCT for short) . For a power function, the distribution of BCT can be directly determined by its boomerang spectrum. In this paper, we investigate the boomerang spectra of two classes power functions over even characteristic finite fields via their differential spectra. The boomerang spectrum of the power function $ {x^{{2^{m+1}} - 1}} $ over $ {\mathbb{F}_{{2^{2m}}}} $ is determined, where $2^{m+1}-1$ is a kind of Niho exponent. The boomerang spectrum of the Gold function $G(x)=x^{2^t+1}$ over $ {\mathbb{F}_{{2^n}}} $ is also determined. It is shown that the Gold function has two-valued boomerang spectrum., Comment: Part of the results of our article are covered by sihem BFA's article

Xiaolei Han,1,2,&ast; Yongxiang Wang,2,3,&ast; Ziying Jiang,1 Yuanjing Li,4 Yi Dong,1 Lin Cong,1– 3 Tingting Hou,1– 3 Yajun Liang,4,5 Erika J Laukka,4 Yifeng Du,1– 3 Chengxuan Qiu1,4 1Department of Neurology, Shandong Provincial Hospital, Shandong University, Jinan, Shandong, People’s Republic of China; 2Department of Neurology, Shandong Provincial Hospital Affiliated to Shandong First Medical University, Jinan, Shandong, People’s Republic of China; 3Shandong Provincial Clinical Research Center for Neurological Diseases, Jinan, Shandong Province, People’s Republic of China; 4Aging Research Center and Center for Alzheimer Research, Department of Neurobiology, Care Sciences and Society, Karolinska Institutet-Stockholm University, Stockholm, Sweden; 5Department of Global Public Health, Karolinska Institutet, Stockholm, Sweden&ast;These authors contributed equally to this workCorrespondence: Yifeng Du, Department of Neurology, Shandong Provincial Hospital, Cheeloo College of Medicine, Shandong University, No. 324 Jingwuweiqi Road, Jinan, Shandong, 250021, People’s Republic of China, Tel +86-531-68776354, Fax +86-531-68773274, Email du-yifeng@hotmail.comPurpose: We explore the associations of individual and composite cardiovascular health metrics with all-cause dementia, Alzheimer’s disease, and vascular dementia among rural-dwelling older adults and the potential age variations in their associations.Patients and Methods: This community-based cross-sectional study included 4980 older adults (age ≥ 65 years; 57.23% women) from the baseline examination of MIND-China. In March–September 2018, data were collected via face-to-face interviews, clinical examinations, and laboratory test. We defined six cardiovascular health metrics according to the modified American Heart Association’s recommendations. We diagnosed dementia and its subtypes following the international criteria. Data were analyzed using logistic regression models.Results: Of all

A smart and flexible textile strain sensor was developed based on a commercial core-spun yarn coated with carbon black (CB) nanoparticles and polydopamine (PDA). The morphology of the textile was studied using scanning electron microscopy (SEM), and the distribution of CB and self polymerized PDA was observed. Thermogravimetric analysis (TGA) revealed the applicable temperature range of the strain sensor. The smart textile showed an outstanding strain sensor response range up to 450%, and its mechanical properties were only slightly reduced as compared to the original textile. The cyclic strain sensor behavior of the smart textile was investigated by stretching the textile 100% for 100 cycles, yielding a stable and durable output electrical signal. Application tests for various human motions were also performed. Finally, two mathematical equations based on tunneling theory were proposed to accurately predict the electrical signal during stretching and to reveal the fundamental properties of the textile. Our findings clearly expand the current knowledge of strain sensors.

Qiuchong Chen,1,&ast; Yixue Zhang,1,&ast; Mengjun Zhang,1 Ziying Li,1 Jindong Liu1,2 1Department of Anesthesiology, The Affiliated Hospital of Xuzhou Medical University, Xuzhou, Jiangsu, People’s Republic of China; 2Jiangsu Province Key Laboratory of Anesthesiology, Xuzhou Medical University, Xuzhou, Jiangsu, People’s Republic of China&ast;These authors contributed equally to this workCorrespondence: Jindong Liu, Department of Anesthesiology, The Affiliated Hospital of Xuzhou Medical University, 99 Huaihai Road West, Quanshan District, Xuzhou, Jiangsu, 221000, People’s Republic of China, Email liujindong1818@163.comObjective: There has been a worldwide increment in acute kidney injury (AKI) incidence among elderly orthopedic operative patients. The AKI prediction model provides patients’ early detection a possibility at risk of AKI; most of the AKI prediction models derive, however, from the cardiothoracic operation. The purpose of this study is to predict the risk of AKI in elderly patients after orthopedic surgery based on machine learning algorithm models.Methods: We organized a retrospective study being comprised of 1000 patients with postoperative AKI undergoing orthopedic surgery from September 2016, to June, 2021. They were divided into training (80%;n=799) and test (20%;n=201) sets.We utilized nine machine learning (ML) algorithms and used intraoperative information and preoperative clinical features to acquire models to predict AKI. The performance of the model was evaluated according to the area under the receiver operating characteristic (AUC), sensitivity, specificity and accuracy. Select the optimal model and establish the nomogram to make the prediction model visualization. The concordance statistic (C-statistic) and calibration curve were used to discriminate and calibrate the nomogram respectively.Results: In predicting AKI, nine ML algorithms posted AUC of 0.656– 1.000 in the training cohort, with the randomforest standing out

NAD(P)H is a key cofactor widely used in biocatalytic reductive transformations, facilitating a wide range of industrially significant reactions which ultimately result in the consumption of the costly cofactor. To make NAD(P)H dependent biotransformations sustainable and economically feasible, different catalytic routes have been investigated to regenerate NAD(P)H. Here we report a comprehensive life cycle assessment (LCA) of these catalytic regeneration methods. Midpoint characterisation and normalisation show that the synthesis of the catalyst, specifically the use of noble metals and energy consumption, dominate the environmental impacts and have the greatest contribution to all considered impact categories. This comparative LCA highlights the need for future investigation into noble metal based catalyst alternatives, to provide cleaner and more sustainable methods of regenerating the cofactor NAD(P)H.

Background: Cardiovascular health (CVH) metrics among Chinese older adults are poorly understood. We investigated sex disparities in CVH metrics and their management among rural-dwelling older adults in China. Methods: This community-based study included 5026 participants (age >= 65years; 57.2% women) in the baseline survey of a multimodal intervention study in rural China. In March-September 2018, data were collected through face-to-face interviews, clinical examinations, and laboratory tests. We defined six CVH metrics (three behavioral factors-smoking, body mass index, and physical activity; three biological factors-blood pressure, total cholesterol, and blood glucose) following the modified American Heart Association's recommendations. We performed descriptive analysis separately for men and women. Results: Of all participants, only 0.8% achieved ideal levels in all six CVH metrics. Men were more likely than women to have ideal levels in all CVH metrics but smoking. Women had higher prevalence of ideal global (9.7% vs. 7.8%) and behavioral (18.3% vs. 9.5%) CVH metrics (p<0.001), whereas men had higher prevalence of ideal biological CVH metrics (5.4% vs. 3.5%, p<0.001). The prevalence of ideal global and behavioral CVH metrics increased with age in both women and men (p for trend<0.001). Women were more likely to be aware of their hypertension and diabetes, and to receive antihypertensive treatment, while men were more likely to achieve the goal of high cholesterol treatment (p<0.05). Conclusions: The CVH metrics among older adults living in the rural communities in China are characterized by an extremely low proportion of optimal global CVH metrics and distinct sex differences, alongside poor management of major biological risk factors.

At non-volcanic passive continental margins, seismic techniques often failed to uniquely define the nature of crustal domains. Here, we overcome this problem by studying the structure and composition of the continent-ocean transition (COT) in the Southwest Sub-basin of the South China Sea, using P and S wave seismic tomography and Vp/Vs ratios, providing unique constraints on lithology. Throughout the image domain, we can rule out large areas of exhumed mantle as Vp/Vs ratios are always <1.9 in the shallow basement layer. Instead, the COT is characterized by extended and fragmented continental crust, and possibly mafic aggregation at the bottom of the crust. In concert with observations from multichannel seismic reflection data, seismic velocities and Vp/Vs ratios suggest that the oldest oceanic crust was formed by starved magmatism, causing rugged basement, thin crust, nearly absent lower crust, and moderately serpentinized mantle below. Our results reveal that rifting occurred without un-roofing continental mantle.

Although research has thoroughly established that employees’ safety citizenship behaviors (SCBs) are critical to workplace safety, less is known about the patterns by which team-level safety stressors affect SCBs. Extending work stress theories to the team level, this study employs a multilevel model and aims to assess two unique mediating mechanisms, felt safety responsibility and affective commitment, through which team safety stressors influence proactive and prosocial safety behaviors respectively. Data were collected from 408 construction workers and their supervisors from 28 project teams in China. Results showed that team safety stressors significantly and negatively predicted both types of SCB. Moreover, felt safety responsibility mediated the relationship between team safety stressors and proactive safety behavior, and affective commitment mediated the relationship between team safety stressors and prosocial safety behavior. This study contributes to workplace safety research by highlighting the important role of team safety stressors in predicting SCBs and different mediating mechanisms for the two types of SCB. Based on our findings, practical interventions aiming at improving workplace safety could be targeted at training managers to provide a supportive work environment where safety roles are clearly and consistently communicated, as well as to attend to potential interpersonal conflicts within the work team. These strategies will encourage more SCBs by promoting workers’ understanding of their responsibilities and enhancing their commitment to the organization.