A deep convolutional neural network for diabetic retinopathy detection via mining local and long-range dependence

Diabetic retinopathy(DR),the main cause of irreversible blindness,is one of the most common complications of diabetes.At present,deep convolutional neural networks have achieved promising performance in automatic DR detection tasks.The convolution operation of methods is a local cross-correlation operation,whose receptive field de-termines the size of the local neighbourhood for processing.However,for retinal fundus photographs,there is not only the local information but also long-distance dependence between the lesion features(e.g.hemorrhages and exudates)scattered throughout the whole image.The proposed method incorporates correlations between long-range patches into the deep learning framework to improve DR detection.Patch-wise re-lationships are used to enhance the local patch features since lesions of DR usually appear as plaques.The Long-Range unit in the proposed network with a residual structure can be flexibly embedded into other trained networks.Extensive experimental results demon-strate that the proposed approach can achieve higher accuracy than existing state-of-the-art models on Messidor and EyePACS datasets.

中国阿尔茨海默病早期预防指南(2024)

阿尔茨海默病是一种病因复杂的严重智力致残疾病,目前尚无有效的治疗方法。我国是世界上老年人口最多、增长最快的国家之一,也是阿尔茨海默病发病率较高的国家。因此,如何早期预防阿尔茨海默病的发生,是现今健康领域最重要的课题之一。我们在国内外相关研究的基础上,通过文献循证、干预研究实践、案例分析、经验总结和专家咨询,结合中国老年健康和社会文化特点,体现中国传统医学、武术、养生以及社区组织优势,制订出具有中国特色的阿尔茨海默病早期预防指南,涵盖了个人、家庭和社会三个层面的预防策略,从积极生活态度、社会活动参与、认知训练、体育锻炼、戒烟限酒、营养、睡眠,血压、血糖、血脂管理,体重和其他慢病管理以及中医开展针对阿尔茨海默病的一级预防。本指南仅供个人、家庭和社会开展阿尔茨海默病预防时参考。

GNAS mutations suppress cell invasion by activating MEG3 in growth hormone–secreting pituitary adenoma

Approximately 30%–40%of growth hormone–secreting pituitary adenomas(GHPAs)harbor somatic activating mutations in GNAS(αsubunit of stimulatory G protein).Mutations in GNAS are associated with clinical features of smaller and less invasive tumors.However,the role of GNAS mutations in the invasiveness of GHPAs is unclear.GNAS mutations were detected in GHPAs using a standard polymerase chain reaction(PCR)sequencing procedure.The expression of mutation-associated maternally expressed gene 3(MEG3)was evaluated with RT-qPCR.MEG3 was manipulated in GH3 cells using a lentiviral expression system.Cell invasion ability was measured using a Transwell assay,and epithelial–mesenchymal transition(EMT)-associated proteins were quantified by immunofluorescence and western blotting.Finally,a tumor cell xenograft mouse model was used to verify the effect of MEG3 on tumor growth and invasiveness.The invasiveness of GHPAs was significantly decreased in mice with mutated GNAS compared with that in mice with wild-type GNAS.Consistently,the invasiveness of mutant GNASexpressing GH3 cells decreased.MEG3 is uniquely expressed at high levels in GHPAs harboring mutated GNAS.Accordingly,MEG3 upregulation inhibited tumor cell invasion,and conversely,MEG3 downregulation increased tumor cell invasion.Mechanistically,GNAS mutations inhibit EMT in GHPAs.MEG3 in mutated GNAS cells prevented cell invasion through the inactivation of the Wnt/β-catenin signaling pathway,which was further validated in vivo.Our data suggest that GNAS mutations may suppress cell invasion in GHPAs by regulating EMT through the activation of the MEG3/Wnt/β-catenin signaling pathway.

Geometric prior guided hybrid deep neural network for facial beauty analysis

Facial beauty analysis is an important topic in human society.It may be used as a guidance for face beautification applications such as cosmetic surgery.Deep neural networks(DNNs)have recently been adopted for facial beauty analysis and have achieved remarkable performance.However,most existing DNN-based models regard facial beauty analysis as a normal classification task.They ignore important prior knowledge in traditional machine learning models which illustrate the significant contribution of the geometric features in facial beauty analysis.To be specific,landmarks of the whole face and facial organs are introduced to extract geometric features to make the decision.Inspired by this,we introduce a novel dual-branch network for facial beauty analysis:one branch takes the Swin Transformer as the backbone to model the full face and global patterns,and another branch focuses on the masked facial organs with the residual network to model the local patterns of certain facial parts.Additionally,the designed multi-scale feature fusion module can further facilitate our network to learn complementary semantic information between the two branches.In model optimisation,we propose a hybrid loss function,where especially geometric regulation is introduced by regressing the facial landmarks and it can force the extracted features to convey facial geometric features.Experiments performed on the SCUT-FBP5500 dataset and the SCUT-FBP dataset demonstrate that our model outperforms the state-of-the-art convolutional neural networks models,which proves the effectiveness of the proposed geometric regularisation and dual-branch structure with the hybrid network.To the best of our knowledge,this is the first study to introduce a Vision Transformer into the facial beauty analysis task.

Anoctamin 4 defines glucose-inhibited neurons in the ventromedial hypothalamus

Glucose is the primary fuel source of the brain,and therefore glucose levels need to be tightly regulated and maintained within a small physiological range.Certainly,the body necessitates a stable supply of energy mainly provided by glucose for various bodily functions.High or low blood glucose levels would impair the physiological functions of various organs of the body.

Dynamic SLAM Visual Odometry Based on Instance Segmentation:A Comprehensive Review

Dynamic Simultaneous Localization and Mapping(SLAM)in visual scenes is currently a major research area in fields such as robot navigation and autonomous driving.However,in the face of complex real-world envi-ronments,current dynamic SLAM systems struggle to achieve precise localization and map construction.With the advancement of deep learning,there has been increasing interest in the development of deep learning-based dynamic SLAM visual odometry in recent years,and more researchers are turning to deep learning techniques to address the challenges of dynamic SLAM.Compared to dynamic SLAM systems based on deep learning methods such as object detection and semantic segmentation,dynamic SLAM systems based on instance segmentation can not only detect dynamic objects in the scene but also distinguish different instances of the same type of object,thereby reducing the impact of dynamic objects on the SLAM system’s positioning.This article not only introduces traditional dynamic SLAM systems based on mathematical models but also provides a comprehensive analysis of existing instance segmentation algorithms and dynamic SLAM systems based on instance segmentation,comparing and summarizing their advantages and disadvantages.Through comparisons on datasets,it is found that instance segmentation-based methods have significant advantages in accuracy and robustness in dynamic environments.However,the real-time performance of instance segmentation algorithms hinders the widespread application of dynamic SLAM systems.In recent years,the rapid development of single-stage instance segmentationmethods has brought hope for the widespread application of dynamic SLAM systems based on instance segmentation.Finally,possible future research directions and improvementmeasures are discussed for reference by relevant professionals.

2023中国阿尔茨海默病数据与防控策略

本报告汇总了阿尔茨海默病及痴呆相关领域现有的最新数据,分析了我国阿尔茨海默病的流行病学、疾病负担、诊断治疗、风险因素、康复护理和疾病筛查等各方面的现状、问题以及趋势,针对目前中国民众和政策制定者最关心的问题,例如:阿尔茨海默病基本数据是怎样的?为什么知晓率相对提高,但是患者就诊愿望相对不高?如何采用有效的措施来应对,等等,提出了目前阿尔茨海默病的防控策略。特别强调了要大力开发经过验证的早期筛查和诊断工具,提供更有效治疗的创新药物,以及鼓励和建立全国性属地化社会互助支持网络。本报告期望能对医学专业人士、患者、家属和照护者、政府政策制定人员、养老机构等有所帮助,能为阿尔茨海默病的防治工作提供支撑,为相关卫生政策的制定提供依据和建议,有助于提高公众对阿尔茨海默病的认识,以期缓解我国阿尔茨海默病疾病的整体负担,推动我国健康老龄化的实现。

硬段结构对含二硫键自修复聚氨酯性能的影响

硬段结构的研究对指导制备兼顾力学性能与自修复性能的聚氨酯弹性体具有重要的意义。文中分别以4,4'-二苯基甲烷二异氰酸酯(MDI)、4,4'-二环己基甲烷二异氰酸酯(HMDI)、甲苯二异氰酸酯(TDI)、异佛尔酮二异氰酸酯(IPDI)及双(2-羟基乙基)二硫醚(HEDS)为硬段部分,聚己内酯二元醇为软段部分,制备了一系列含脂肪族二硫键的自修复聚氨酯,探究了不同异氰酸酯、硬段含量和扩链剂对其力学性能与自修复性能的影响。结果表明,对于异氰酸酯结构来讲,脂环族异氰酸酯制备的自修复聚氨酯具备更好的力学性能与自修复性能;对于脂环族异氰酸酯体系聚氨酯的硬段含量而言,一定范围越高,二硫键与氢键的含量也越高,越有利于材料的力学性能与自修复性能;对于扩链剂结构来讲,脂肪族二硫键主要是促进较高温度条件下的自修复性能,对于室温自修复性能起到的作用有限。

Two-view attention-guided convolutional neural network for mammographic image classification

Deep learning has been widely used in the field of mammographic image classification owing to its superiority in automatic feature extraction.However,general deep learning models cannot achieve very satisfactory classification results on mammographic images because these models are not specifically designed for mammographic images and do not take the specific traits of these images into account.To exploit the essential discriminant information of mammographic images,we propose a novel classification method based on a convolutional neural network.Specifically,the proposed method designs two branches to extract the discriminative features from mammographic images from the mediolateral oblique and craniocaudal(CC)mammographic views.The features extracted from the two-view mammographic images contain complementary information that enables breast cancer to be more easily distinguished.Moreover,the attention block is introduced to capture the channel-wise information by adjusting the weight of each feature map,which is beneficial to emphasising the important features of mammographic images.Furthermore,we add a penalty term based on the fuzzy cluster algorithm to the cross-entropy function,which improves the generalisation ability of the classification model by maximising the interclass distance and minimising the intraclass distance of the samples.The experimental results on The Digital database for Screening Mammography INbreast and MIAS mammography databases illustrate that the proposed method achieves the best classification performance and is more robust than the compared state-ofthe-art classification methods.

Twin model-based fault detection and tolerance approach for in-core self-powered neutron detectors

The in-core self-powered neutron detector(SPND)acts as a key measuring device for the monitoring of parameters and evaluation of the operating conditions of nuclear reactors.Prompt detection and tolerance of faulty SPNDs are indispensable for reliable reactor management.To completely extract the correlated state information of SPNDs,we constructed a twin model based on a generalized regression neural network(GRNN)that represents the common relationships among overall signals.Faulty SPNDs were determined because of the functional concordance of the twin model and real monitoring sys-tems,which calculated the error probability distribution between the model outputs and real values.Fault detection follows a tolerance phase to reinforce the stability of the twin model in the case of massive failures.A weighted K-nearest neighbor model was employed to reasonably reconstruct the values of the faulty signals and guarantee data purity.The experimental evaluation of the proposed method showed promising results,with excellent output consistency and high detection accuracy for both single-and multiple-point faulty SPNDs.For unexpected excessive failures,the proposed tolerance approach can efficiently repair fault behaviors and enhance the prediction performance of the twin model.

Multi-Robot Privacy-Preserving Algorithms Based on Federated Learning:A Review

The robotics industry has seen rapid development in recent years due to the Corona Virus Disease 2019.With the development of sensors and smart devices,factories and enterprises have accumulated a large amount of data in their daily production,which creates extremely favorable conditions for robots to perform machine learning.However,in recent years,people’s awareness of data privacy has been increasing,leading to the inability to circulate data between different enterprises,resulting in the emergence of data silos.The emergence of federated learning provides a feasible solution to this problem,and the combination of federated learning and multi-robot systems can break down data silos and improve the overall performance of robots.However,as scholars have studied more deeply,they found that federated learning has very limited privacy protection.Therefore,how to protect data privacy from infringement remains an important issue.In this paper,we first give a brief introduction to the current development of multi-robot and federated learning;second,we review three aspects of privacy protection methods commonly used,privacy protection methods for multi-robot,and Other Problems Faced by Multi-robot Systems,focusing on method comparisons and challenges;and finally draw conclusions and predict possible future research directions.

Performance comparison of lithium fractionation from magnesium via continuous selective nanofiltration/electrodialysis

Although selective nanofiltration(SNF)and selective electrodialysis(SED)have been widely adopted in the field of Mg^(2+)/Li^(+)separation,their differences have not been illustrated systematically.In this study,for the first time,SNF and SED processes in continuous mode were studied for Li+fractionation from the same brine with high Mg/Li ratios and their differences were discussed in detail.For a fair analysis of the two processes,typical factors were optimized.Specifically,the optimal operating pressure and feed flow rate for SNF were 2.4 MPa and 140 L·h^(-1),respectively,while the optimal cell-pair voltage and replenishment flow rate for SED were 1.0 V and 14 L·h^(-1),respectively.Although the Li^(+)fractionation capacity of the two processes were similar,the selectivity coefficient of SNF was 24.7% higher than that of SED and,thus,the Mg/Li ratio in purified stream of the former was 19.0% lower than that of the latter.Due to higher ion driving force,SED had clear advantages in recovery ratio and concentration effects.Meanwhile,the specific energy consumption of SED was 20.1% lower than that of SNF.This study provided a better understanding and guidance for the application and improvement of the two technologies.

Water-Based Environmentally Friendly Pesticide Formulations Based on Cyclodextrin/Pesticide Loading System

Difenoconazole(DIF)is a representative variety of broad-spectrum triazole fungicides and liposoluble pesticides.However,the water solubility of DIF is so poor that its application is limited in plant protection.In addition,the conventional formulations of DIF always contain abundant organic solvents,which may cause pollution of the environment.In this study,two DIF/cyclodextrins(CDs)inclusion complexes(ICs)were successfully prepared,which were DIF/β-CD IC and DIF/hydroxypropyl-β-CD IC(DIF/HP-β-CD IC).The effect of cyclodextrins on the water solubility and the antifungal effect of liposoluble DIF pesticide were investigated.According to the phase solubility test,the molar ratio and apparent stability constant of ICs were obtained.Fourier transform infrared spectroscopy,thermal gravity analysis,X-ray diffraction and scanning electron microscopy were used systematically to characterize the formation and characteristics of ICs.The results noted that DIF successfully entered the cavities of two CDs.In addition,the antifungal effect test proved the better performance of DIF/HP-β-CD IC,which exceeded that of DIF emulsifiable concentrate.Therefore,our study provides informative direction for the intelligent use of liposoluble pesticides with cyclodextrins to develop water-based environmentally friendly formulations.

Ultrafast antiferromagnet rearrangement in Co/IrMn/CoGd trilayers

Antiferromagnets offer great potential for high-speed data processing applications,as they can expend spintronic devices from a static storage and gigahertz frequency range to the terahertz range.However,their zero net magnetization makes them difficult to manipulate and detect.In recent years,there has been a lot of attention given to the ultrafast manipulation of magnetic order using ultra-short single laser pulses,but it remains unknown whether a similar scenario can be observed in antiferromagnets.In this work,we demonstrate the manipulation of antiferromagnets with a single femtosecond laser pulse in perpendicular exchange-biased Co/Ir Mn/Co Gd trilayers.We study the dual exchange bias interlayer interaction in quasi-static conditions and competition in ultrafast antiferromagnet rearrangement.Our results show that,compared to conventional ferromagnetic/antiferromagnetic systems,the Ir Mn antiferromagnet can be ultrafast and efficiently manipulated by the coupled Co Gd ferrimagnetic layer,which paves the way for potential energy-efficient spintronic devices.

Visual SLAM Based on Object Detection Network:A Review

Visual simultaneous localization and mapping(SLAM)is crucial in robotics and autonomous driving.However,traditional visual SLAM faces challenges in dynamic environments.To address this issue,researchers have proposed semantic SLAM,which combines object detection,semantic segmentation,instance segmentation,and visual SLAM.Despite the growing body of literature on semantic SLAM,there is currently a lack of comprehensive research on the integration of object detection and visual SLAM.Therefore,this study aims to gather information from multiple databases and review relevant literature using specific keywords.It focuses on visual SLAM based on object detection,covering different aspects.Firstly,it discusses the current research status and challenges in this field,highlighting methods for incorporating semantic information from object detection networks into mileage measurement,closed-loop detection,and map construction.It also compares the characteristics and performance of various visual SLAM object detection algorithms.Lastly,it provides an outlook on future research directions and emerging trends in visual SLAM.Research has shown that visual SLAM based on object detection has significant improvements compared to traditional SLAM in dynamic point removal,data association,point cloud segmentation,and other technologies.It can improve the robustness and accuracy of the entire SLAM system and can run in real time.With the continuous optimization of algorithms and the improvement of hardware level,object visual SLAM has great potential for development.

Path integral solutions for n-dimensional stochastic differential equations underα-stable Lévy excitation

In this paper,the path integral solutions for a general n-dimensional stochastic differential equa-tions(SDEs)withα-stable Lévy noise are derived and verified.Firstly,the governing equations for the solutions of n-dimensional SDEs under the excitation ofα-stable Lévy noise are obtained through the characteristic function of stochastic processes.Then,the short-time transition probability density func-tion of the path integral solution is derived based on the Chapman-Kolmogorov-Smoluchowski(CKS)equation and the characteristic function,and its correctness is demonstrated by proving that it satis-fies the governing equation of the solution of the SDE,which is also called the Fokker-Planck-Kolmogorov equation.Besides,illustrative examples are numerically considered for highlighting the feasibility of the proposed path integral method,and the pertinent Monte Carlo solution is also calculated to show its correctness and effectiveness.

An experimental study on fire characteristics of urban complex underground space based on BIM

Compared with the space on the ground,if there is a fire in the urban complex underground space,the loss will be greatly harmful.In addition,the complex underground space is usually connected with other large space areas and densely populated.Once a fire occurs in the complex underground space,it will cause huge property losses and casualties.In order to reduce the risk of fire,it is necessary to deeply understand the development rules and characteristics of fire in the complex underground space of the city.This article has mainly carried on the following work:(I)A particularly complex model of the multi‐storey subway station was built.On this basis,three groups of comparative experiments were conducted to study the effects of fire power,fire location and smoke control system on fire development,and the conclusion that fire location is the most important factor for fire development was obtained;(II)In order to explore the entire space fire and the local space fire,CFD(Computational Fluid Dynamics)is used to build a large‐size fire model and a small‐size fire model respectively;(III)Multiple detector data as temperature slices were built,and it is expected to make full use of the simulation data to deduce the important index of fire location in the early stage of fire.All of the works in this paper will provide reference experimental data for the prevention and firefighting of a sudden fire in the complex underground space.

Advance on stochastic dynamics/nonlinear and stochastic dynamics

Nonlinear and stochastic dynamics is essential and practical in almost all branches of natural science and engineering.It has been a central subject to understand various complex dynamics,such as random vibration,stochastic transition,synchronization,et al.,in the areas of mechanical and aerospace engineering,physics and chemistry.For example,the stochastic resonance has been utilized effectively in mechanical fault diagnosis and weak signal detection.Naturally,many methods have been developed to get the stochastic responses,including stochastic averaging method.

Background modeling methods in video analysis： A review and comparative evaluation

Foreground detection methods can be applied to efficiently distinguish foreground objects including moving or static objects from back- ground which is very important in the application of video analysis, especially video surveillance. An excellent background model can obtain a good foreground detection results. A lot of background modeling methods had been proposed, but few comprehensive evaluations of them are available. These methods suffer from various challenges such as illumination changes and dynamic background. This paper first analyzed advantages and disadvantages of various background modeling methods in video analysis applications and then compared their performance in terms of quality and the computational cost. The Change detection.Net （CDnet2014） dataset and another video dataset with different envi- ronmental conditions （indoor, outdoor, snow） were used to test each method. The experimental results sufficiently demonstrated the strengths and drawbacks of traditional and recently proposed state-of-the-art background modeling methods. This work is helpful for both researchers and engineering practitioners. Codes of background modeling methods evaluated in this paper are available atwww.yongxu.org/lunwen.html.

2022年中国阿尔茨海默病知晓与需求现状调查

目的:为了更加全面了解我国老年性痴呆(阿尔茨海默病)防治工作的现状及发展趋势,更加精准地为防治工作提供政策咨询及对策建议,开展了本次阿尔茨海默病知晓与需求现状调查。对象与方法:通过线上问卷,对18岁以上普通成年居民进行调查,居民来自全国各个省市自治区和港澳台地区,调查总样本为20671人。结果:被调查人群阿尔茨海默病的总知晓率为95.9%,经统计学标化后为91.5%。不同人群对阿尔茨海默病的总知晓率存在一定差异;认为阿尔茨海默病或老年痴呆病可以预防的比例是87.9%;被调查居民觉得居家照护患者遇到的困难最主要表现在精力不够、照护能力不足和缺少专业的失智养老服务机构;患者的最主要的照护者是配偶,占47.9%,对阿尔茨海默病康复治疗内容了解的仅占23.3%;对合理安排患者营养知晓率为39%,在被调查人群中,主动愿意去医院就诊的比例只有12.9%,其中农村居民和低学历居民主动就诊愿望较低。结论:本次调查结果为更加精准地制定阿尔茨海默病防治对策,提供了重要的科学依据。