Evolutionary Neural Architecture Search and Its Applications in Healthcare

Most of the neural network architectures are based on human experience,which requires a long and tedious trial-and-error process.Neural architecture search(NAS)attempts to detect effective architectures without human intervention.Evolutionary algorithms(EAs)for NAS can find better solutions than human-designed architectures by exploring a large search space for possible architectures.Using multiobjective EAs for NAS,optimal neural architectures that meet various performance criteria can be explored and discovered efficiently.Furthermore,hardware-accelerated NAS methods can improve the efficiency of the NAS.While existing reviews have mainly focused on different strategies to complete NAS,a few studies have explored the use of EAs for NAS.In this paper,we summarize and explore the use of EAs for NAS,as well as large-scale multiobjective optimization strategies and hardware-accelerated NAS methods.NAS performs well in healthcare applications,such as medical image analysis,classification of disease diagnosis,and health monitoring.EAs for NAS can automate the search process and optimize multiple objectives simultaneously in a given healthcare task.Deep neural network has been successfully used in healthcare,but it lacks interpretability.Medical data is highly sensitive,and privacy leaks are frequently reported in the healthcare industry.To solve these problems,in healthcare,we propose an interpretable neuroevolution framework based on federated learning to address search efficiency and privacy protection.Moreover,we also point out future research directions for evolutionary NAS.Overall,for researchers who want to use EAs to optimize NNs in healthcare,we analyze the advantages and disadvantages of doing so to provide detailed guidance,and propose an interpretable privacy-preserving framework for healthcare applications.

Direct recycling of Li-ion batteries from cell to pack level:Challenges and prospects on technology,scalability,sustainability,and economics

Direct recycling is a novel approach to overcoming the drawbacks of conventional lithium-ion battery(LIB)recycling processes and has gained considerable attention from the academic and industrial sectors in recent years.The primary objective of directly recycling LIBs is to efficiently recover and restore the active electrode materials and other components in the solid phase while retaining electrochemical performance.This technology's advantages over traditional pyrometallurgy and hydrometallurgy are costeffectiveness,energy efficiency,and sustainability,and it preserves the material structure and morphology and can shorten the overall recycling path.This review extensively discusses the advancements in the direct recycling of LIBs,including battery sorting,pretreatment processes,separation of cathode and anode materials,and regeneration and quality enhancement of electrode materials.It encompasses various approaches to successfully regenerate high-value electrode materials and streamlining the recovery process without compromising their electrochemical properties.Furthermore,we highlight key challenges in direct recycling when scaled from lab to industries in four perspectives:(1)battery design,(2)disassembling,(3)electrode delamination,and(4)commercialization and sustainability.Based on these challenges and changing market trends,a few strategies are discussed to aid direct recycling efforts,such as binders,electrolyte selection,and alternative battery designs;and recent transitions and technological advancements in the battery industry are presented.

A Robust Framework for Multimodal Sentiment Analysis with Noisy Labels Generated from Distributed Data Annotation

Multimodal sentiment analysis utilizes multimodal data such as text,facial expressions and voice to detect people’s attitudes.With the advent of distributed data collection and annotation,we can easily obtain and share such multimodal data.However,due to professional discrepancies among annotators and lax quality control,noisy labels might be introduced.Recent research suggests that deep neural networks(DNNs)will overfit noisy labels,leading to the poor performance of the DNNs.To address this challenging problem,we present a Multimodal Robust Meta Learning framework(MRML)for multimodal sentiment analysis to resist noisy labels and correlate distinct modalities simultaneously.Specifically,we propose a two-layer fusion net to deeply fuse different modalities and improve the quality of the multimodal data features for label correction and network training.Besides,a multiple meta-learner(label corrector)strategy is proposed to enhance the label correction approach and prevent models from overfitting to noisy labels.We conducted experiments on three popular multimodal datasets to verify the superiority of ourmethod by comparing it with four baselines.

MXenes在水系锌离子电池中的应用研究进展

水系锌离子电池(AZIBs)作为一种低成本、高安全的新兴且前景广阔的储能技术近年来备受关注。新型MXenes材料由于其独特的结构特征和物理化学性质,如易调节的二维结构、优异的导电性、化学组成多样和可控的表面化学特性,在AZIBs中表现出独特的应用优势。本文全面综述近年来MXenes在AZIBs中应用的研究进展,探讨MXenes应用于AZIBs正负极的结构设计及性能优化策略:在正极方面,MXenes可直接作为活性物质或活性物质前驱体、基体材料,以获得高活性、优异的循环寿命和倍率性能;在负极方面,MXenes可作为锌沉积的二维/三维载体、亲锌基体及锌金属界面保护层,以减缓电化学反应过程中锌金属的腐蚀和枝晶生长。此外,本文也对MXenes基材料在AZIBs中应用的发展方向进行展望。

Exosome-transported IncRNA H19 regulates insulin-like growth factor-1 via the H19/let-7a/insulin-like growth factor-1 receptor axis in ischemic stroke

LncRNA(long non-coding RNA) H19 is a transcript of the H19 gene that is expressed during embryogenesis.We previously discove red a role for circular lncRNA H19 in the onset and prognosis of cerebral ischemic stroke.In this study,we used serum from patients with ischemic stroke,and mouse and cell culture models to elucidate the roles of plasma and neuronal exosomes in the regulatory effect of lncRNA H19 on insulin-like growth factor-1 and its mechanism in ischemic stroke,using western blotting,quantitative real-time polymerase chain reaction,and enzyme-linked immunosorbent assays.Plasma exosomal IncRNA H19 was negatively associated with blood levels of insulin-like growth factor-1 in samples from patients with cerebral ischemic stroke.In a mouse model,levels of exosomal IncRNA H19 were positively correlated with plasma and cerebral lncRNA H19.In a cell co-culture model,we confirmed that IncRNA H19 was transported from neuro ns to astrocytes by exosomes to induce downregulation of insulin-like growth factor-1 through the H19/let-7 a/insulin-like growth factor-1 receptor axis.This study provides the first evidence for the transpo rtation of IncRNA H19 by exosomes and the relationship between IncRNA H19 and insulinlike growth factor-1.

Opportunistic admission and resource allocation for slicing enhanced IoT networks

Network slicing is envisioned as one of the key techniques to meet the extremely diversified service requirements of the Internet of Things(IoT)as it provides an enhanced user experience and elastic resource configuration.In the context of slicing enhanced IoT networks,both the Service Provider(SP)and Infrastructure Provider(InP)face challenges of ensuring efficient slice construction and high profit in dynamic environments.These challenges arise from randomly generated and departed slice requests from end-users,uncertain resource availability,and multidimensional resource allocation.Admission and resource allocation for distinct demands of slice requests are the key issues in addressing these challenges and should be handled effectively in dynamic environments.To this end,we propose an Opportunistic Admission and Resource allocation(OAR)policy to deal with the issues of random slicing requests,uncertain resource availability,and heterogeneous multi-resources.The key idea of OAR is to allow the SP to decide whether to accept slice requests immediately or defer them according to the load and price of resources.To cope with the random slice requests and uncertain resource availability,we formulated this issue as a Markov Decision Process(MDP)to obtain the optimal admission policy,with the aim of maximizing the system reward.Furthermore,the buyer-seller game theory approach was adopted to realize the optimal resource allocation,while motivating each SP and InP to maximize their rewards.Our numerical results show that the proposed OAR policy can make reasonable decisions effectively and steadily,and outperforms the baseline schemes in terms of the system reward.

Stochastic programming based multi-arm bandit offloading strategy for internet of things

In order to solve the high latency of traditional cloud computing and the processing capacity limitation of Internet of Things(IoT)users,Multi-access Edge Computing(MEC)migrates computing and storage capabilities from the remote data center to the edge of network,providing users with computation services quickly and directly.In this paper,we investigate the impact of the randomness caused by the movement of the IoT user on decision-making for offloading,where the connection between the IoT user and the MEC servers is uncertain.This uncertainty would be the main obstacle to assign the task accurately.Consequently,if the assigned task cannot match well with the real connection time,a migration(connection time is not enough to process)would be caused.In order to address the impact of this uncertainty,we formulate the offloading decision as an optimization problem considering the transmission,computation and migration.With the help of Stochastic Programming(SP),we use the posteriori recourse to compensate for inaccurate predictions.Meanwhile,in heterogeneous networks,considering multiple candidate MEC servers could be selected simultaneously due to overlapping,we also introduce the Multi-Arm Bandit(MAB)theory for MEC selection.The extensive simulations validate the improvement and effectiveness of the proposed SP-based Multi-arm bandit Method(SMM)for offloading in terms of reward,cost,energy consumption and delay.The results showthat SMMcan achieve about 20%improvement compared with the traditional offloading method that does not consider the randomness,and it also outperforms the existing SP/MAB based method for offloading.

Experimental Study of Microalgae Cultivation under Selective Illumination by Ag/CoSO_(4)for Bioelectrode Materials Preparation

Microalgae biomass is an ideal precursor to prepare renewable carbon materials,which has broad application.The bioaccumulation efficiency(lipids,proteins,carbohydrates)and biomass productivity of microalgae are influenced by spectroscopy during the culture process.In this study,a bilayer plate-type photobioreactor was designed to cultivate Chlorella protothecoides with spectral selectivity by nanofluids.Compared to culture without spectral selectivity,the spectral selectivity of Ag/CoSO_(4)nanofluids increased microalgae biomass by 5.76%,and the spectral selectivity of CoSO_(4)solution increased by 17.14%.In addition,the spectral selectivity of Ag/CoSO_(4)nanofluids was more conducive to the accumulation of nutrients(29.46%lipids,50.66%proteins,and 17.86%carbohydrates)in microalgae.Further cultured chlorella was utilized to prepare bioelectrode materials,it was found that algal based biochar had a good pore structure(micro specific surface area:1627.5314 m^(2)/g,average pore size:0.21294 nm).As the current density was 1 A/g,the specific capacitance reached 230 F/g,appearing good electrochemical performance.

基于远程诊断系统的糖尿病眼病综合防治服务模式实践探讨

目的:总结浦东新区糖尿病眼病综合防治服务模式的实施成效,为建立完善以社区为基础的糖尿病眼病综合防治服务模式提供实践参考。方法:于2018-08/2019-08将浦东新区45家社区纳入慢性病管理的糖尿病患者作为筛查对象,根据年龄、性别进行分层抽样,对纳入对象的远程筛查眼健康数据进行分析,并通过问卷调查掌握满意度及参与意愿。结果:本次筛查纳入DM患者39055例,均首次建立眼健康档案,远程诊断初步确诊为糖尿病视网膜病变(DR)者15.870%(6198/39055),其中轻度非增殖性DR、中度非增殖性DR、重度非增殖性DR及增殖性DR者分别为3.974%、9.904%、1.746%、0.246%;需转诊率44.325%(17311/39055),实际复诊率5.846%(1012/17311)。此外,DM患者远程筛查满意度与参与意愿均在85%以上。结论:基于远程诊断系统的糖尿病眼病综合防治服务模式对于实现糖尿病眼病早发现、早诊断、早干预是可行的,但在实际应用中还需进一步优化与完善。

蒺藜皂苷通过调控Wnt3a/β-catenin信号通路抑制胃癌细胞增殖、迁移侵袭的机制研究

背景蒺藜皂苷是从中药蒺藜内提取的活性成分,具有抗癌、抗炎、调节免疫、等多种药理活性,已有研究表明蒺藜皂苷能抑制胃癌细胞生长和诱导凋亡,本研究主要探究蒺藜皂苷可能抑制胃癌细胞增殖、迁移和侵袭.目的探讨蒺藜皂苷通过调控Wnt3a/β-catenin信号通路抑制胃癌细胞增殖、迁移侵袭的机制研究.方法以胃癌细胞HGC-27为体外研究对象,用浓度0、20 mg/L、40 mg/L蒺藜皂苷进行处理,记为对照组(Control)、蒺藜皂苷20 mg/L组、蒺藜皂苷40 mg/L组;将siRNA NC、siRNA Wnt3a转染胃癌细胞中,用40 mg/L蒺藜皂苷处理,记为蒺藜皂苷+siRNA NC组、蒺藜皂苷+siRNA Wnt3a组.用噻唑蓝(methylthiazolyldiphenyl-tetrazolium bromide,MTT)检测细胞增殖活性;伤口愈合实验检测细胞迁移;Transwell小室检测细胞侵袭;蛋白免疫印迹法(Western blot)检测p21、细胞增殖核抗原-67(proliferating nuclear antigen-67,Ki67)、上皮性钙黏附素(epithelical cadherin,E-cadherin)、神经性钙黏附素(neural cadherin,N-cadherin)、波形蛋白(Vimentin)、Wnt3a、β-catenin蛋白表达.结果与Control组比较,蒺藜皂苷20 mg/L组、蒺藜皂苷40 mg/L组胃癌细胞活性、迁移率、侵袭细胞数显著降低,p21、E-cadherin蛋白表达增加,Ki67、N-cadherin、Vimentin、Wnt3a、β-catenin蛋白表达降低.与蒺藜皂苷+siRNA NC组组相比,蒺藜皂苷+siRNA Wnt3a组胃癌细胞活性、迁移率、侵袭细胞数显著降低,Wnt3a、β-catenin、Ki67、N-cadherin、Vimentin蛋白表达降低,p21、E-cadherin蛋白表达增加.结论蒺藜皂苷可能通过抑制Wnt3a/β-catenin信号通路抑制胃癌细胞增殖、迁移、侵袭和上皮间质转化.

钠离子电池炭基负极材料研究进展

钠离子电池是目前新兴的低成本储能技术,因在大规模电化学储能中具有较好的应用前景而受到了国内外学者广泛的关注与研究。作为钠离子电池的关键电极材料之一,非石墨的炭质材料因具有储钠活性高、成本低廉、无毒无害等诸多优点,而被认为是钠离子电池实际应用时负极的最佳选择。本文详细综述了目前钠离子电池炭基负极材料的研究进展,重点介绍了炭质材料的储钠机理与特性,分析了炭材料结构与电化学性能之间的关系,探讨了其存在的问题,为钠离子电池炭基负极材料的发展提供有益的认识。

Electrostatic Self-assembly of 0D–2D SnO_(2) Quantum Dots/Ti_(3)C_(2)T_(x) MXene Hybrids as Anode for Lithium-Ion Batteries

MXenes,a new family of two-dimensional(2D)materials with excellent electronic conductivity and hydrophilicity,have shown distinctive advantages as a highly conductive matrix material for lithium-ion battery anodes.Herein,a facile electrostatic self-assembly of SnO2 quantum dots(QDs)on Ti3C2Tx MXene sheets is proposed.The as-prepared SnO2/MXene hybrids have a unique 0D-2D structure,in which the 0D SnO2 QDs(~4.7 nm)are uniformly distributed over 2D Ti3C2Tx MXene sheets with controllable loading amount.The SnO2 QDs serve as a high capacity provider and the“spacer”to prevent the MXene sheets from restacking;the highly conductive Ti3C2Tx MXene can not only provide efficient pathways for fast transport of electrons and Li ions,but also buffer the volume change of SnO2 during lithiation/delithiation by confining SnO2 QDs between the MXene nanosheets.Therefore,the 0D-2D SnO2 QDs/MXene hybrids deliver superior lithium storage properties with high capacity(887.4 mAh g?1 at 50 mA g?1),stable cycle performance(659.8 mAh g?1 at 100 mA g?1 after 100 cycles with a capacity retention of 91%)and excellent rate performance(364 mAh g?1 at 3 A g?1),making it a promising anode material for lithium-ion batteries.

Performance analysis and comparison of PoW,PoS and DAG based blockchains

In the blockchain,the consensus mechanism plays a key role in maintaining the security and legitimation of contents recorded in the blocks.Various blockchain consensus mechanisms have been proposed.However,there is no technical analysis and comparison as a guideline to determine which type of consensus mechanism should be adopted in a specific scenario/application.To this end,this work investigates three mainstream consensus mechanisms in the blockchain,namely,Proof of Work(PoW),Proof of Stake(PoS),and Direct Acyclic Graph(DAG),and identifies their performances in terms of the average time to generate a new block,the confirmation delay,the Transaction Per Second(TPS)and the confirmation failure probability.The results show that the consensus process is affected by both network resource(computation power/coin age,buffer size)and network load conditions.In addition,it shows that PoW and PoS are more sensitive to the change of network resource while DAG is more sensitive to network load conditions.

MOF-Derived ZnS Nanodots/Ti_(3)C_(2)T_(x) MXene Hybrids Boosting Superior Lithium Storage Performance

ZnS has great potentials as an anode for lithium storage because of its high theoretical capacity and resource abundance;however,the large volume expansion accompanied with structural collapse and low conductivity of ZnS cause severe capacity fading and inferior rate capability during lithium storage. Herein,0D-2 D ZnS nanodots/Ti_(3)C_(2)T_x MXene hybrids are prepared by anchoring ZnS nanodots on Ti_(3)C_(2)T_(x) MXene nanosheets through coordination modulation between MXene and MOF precursor(ZIF-8) followed with sulfidation. The MXene substratecoupled with the ZnS nanodots can synergistically accommodate volume variation of ZnS over charge–discharge to realize stable cyclability. As revealed by XPS characterizations and DFT calculations,the strong interfacial interaction between ZnS nanodots and MXene nanosheets can boost fast electron/lithium-ion transfer to achieve excellent electrochemical activity and kinetics for lithium storage. Thereby,the as-prepared ZnS nanodots/MXene hybrid exhibits a high capacity of 726.8 mAh g^(-1) at 30 mA g^(-1),superior cyclic stability(462.8 mAh g^(-1) after 1000 cycles at 0.5 A g^(-1)),and excellent rate performance. The present results provide new insights into the understanding of the lithium storage mechanism of ZnS and the revealing of the e ects of interfacial interaction on lithium storage performance enhancement.

Blockchain-enabled resource management and sharing for 6G communications

The sixth-generation(6G)network must provide better performance than previous generations to meet the requirements of emerging services and applications,such as multi-gigabit transmission rate,higher reliability,and sub-1 ms latency and ubiquitous connection for the Internet of Everything(IoE).However,with the scarcity of spectrum resources,efficient resource management and sharing are crucial to achieving all these ambitious requirements.One possible technology to achieve all this is the blockchain.Because of its inherent properties,the blockchain has recently gained an important position,which is of great significance to the 6G network and other networks.In particular,the integration of the blockchain in 6G will enable the network to monitor and manage resource utilization and sharing efficiently.Hence,in this paper,we discuss the potentials of the blockchain for resource management and sharing in 6G using multiple application scenarios,namely,Internet of things,deviceto-device communications,network slicing,and inter-domain blockchain ecosystems.

An intelligent task offloading algorithm(iTOA)for UAV edge computing network

Unmanned Aerial Vehicle(UAV)has emerged as a promising technology for the support of human activities,such as target tracking,disaster rescue,and surveillance.However,these tasks require a large computation load of image or video processing,which imposes enormous pressure on the UAV computation platform.To solve this issue,in this work,we propose an intelligent Task Offloading Algorithm(iTOA)for UAV edge computing network.Compared with existing methods,iTOA is able to perceive the network’s environment intelligently to decide the offloading action based on deep Monte Calor Tree Search(MCTS),the core algorithm of Alpha Go.MCTS will simulate the offloading decision trajectories to acquire the best decision by maximizing the reward,such as lowest latency or power consumption.To accelerate the search convergence of MCTS,we also proposed a splitting Deep Neural Network(sDNN)to supply the prior probability for MCTS.The sDNN is trained by a self-supervised learning manager.Here,the training data set is obtained from iTOA itself as its own teacher.Compared with game theory and greedy search-based methods,the proposed iTOA improves service latency performance by 33%and 60%,respectively.

Unintentional forking analysis in wireless blockchain networks

The forking problem plays a key role in the security issue,which is a major concern in the blockchain system.Although many works studied the attack strategy,consensus mechanism,privacy-protecting and security performance analysis,most of them only address the intentional forking caused by a malicious attacker.In fact,without any attacker,unintentional forking still remains due to transmission delay and failure,especially in wireless network scenarios.To this end,this paper investigates the reason for generating unintentional forking and derives the forking probability expression in Wireless Blockchain Networks(WBN).Furthermore,in order to illustrate the unintentional forking on the blockchain system,the performances in terms of resource utilization rate,block generation time,and Transaction Per Second(TPS)are investigated.The numerical results show that the target difficulty of hash algorithm in generating a new block,the delay time of broadcasting,the network scale,and the transmission failure probability would affect the unintentional forking probability significantly,which can provide a reliable basis for avoiding forking to save resource consumption and improving system performance.

Scleral TGF-β1 and Smad3 expression is altered by TCM Bu Jing Yi Shi Tablets in guinea pigs with form-deprivation myopia

Objective:To explore whether the traditional Chinese medicine(TCM)Bu Jing Yi Shi Tablets alters the expression of scleral TGF-b1 and Smad3 in guinea pigs with formdeprivation myopia(FDM).Methods:Sixty-five guinea pigs were randomly divided into control,model,low-,medium-,and high-dose treatment groups.Except for the control group,FDM was induced by covering the right eye of each animal with opaque latex.The treatment groups were gavaged with different suspension concentrations of Bu Jing Yi Shi Tablets.Refraction and axial length were performed before and after myopia induction.At the end of the experiment,all right eyes were extracted,and scleral sections were prepared for staining and TGF-b1 and Smad3 immunohistochemistry.Scleral thickness and area,the scleral fibroblast quantity,and scleral TGFb1 and Smad3 expressions were measured.Results:The 5 FDM groups had the same initial axial length and diopter,the final diopter and axial length of the model group were significantly increased compared with the control group and treatment groups(P<.01).The axial length of each treatment group decreased as the dose decreased compared with the model group(P<.01);the total scleral area(P<.05 e.01)and the number of scleral fibroblasts(P<.01)in the model group were significantly lower than the treatment groups.Both the TGF-b1 and Smad3 integral optical densities in the model group were significantly lower than the control and medium-and high-dose treatment groups(P<.01).TGF-b1 and Smad3 mRNAs in the model group were decreased compared with the control group,but increased in expression after treatment.

Caching resource management of mobile edge network based on Stackelberg game

Mobile edge caching technology is gaining more and more attention because it can effectively improve the Quality of Experience (QoE) of users and reduce backhaul burden. This paper aims to improve the utility of mobile edge caching technology from the perspectie of caching resource management by examining a network composed of one operator, multiple users and Content Providers (CPs). The caching resource management model is constructed on the premise of fully considering the QoE of users and the servicing capability of the Base Station (BS). In order to create the best caching resource allocation scheme, the original problem is transformed into a multi-leader multi-follower Stackelberg game model through the analysis of the system model. The strategy combinations and the utility functions of players are analyzed. The existence and uniqueness of the Nash Equilibrium (NE) solution are also analyzed and proved. The optimal strategy combinations and the best responses are deduced in detail. Simulation results and analysis show that the proposed model and algorithm can achieve the optimal allocation of caching resource and improve the QoE of users.

Study of the tungsten sputtering source suppression by wall conditionings in the EAST tokamak

The steady fusion plasma operation is constrained by tungsten(W)material sputtering issue in the EAST tokamak.In this work,the suppression of W sputtering source has been studied by advanced wall conditionings.It is also concluded that the W sputtering yield becomes more with increasing carbon(C)content in the main deuterium(D)plasma.In EAST,the integrated use of discharge cleanings and lithium(Li)coating has positive effects on the suppression of W sputtering source.In the plasma recovery experiments,it is suggested that the W intensity is reduced by approximately 60%with the help of~35 h Ion Cyclotron Radio Frequency Discharge Cleaning(ICRF-DC)and~40 g Li coating after vacuum failure.The first wall covered by Li film could be relieved from the bombardment of energetic particles,and the impurity in the vessel would be removed through the particle induced desorption and isotope exchange during the discharge cleanings.In general,the sputtering yield of W would decrease from the source,on the bias of the improvement of wall condition and the mitigation of plasmawall interaction process.It lays important base of the achievement of high-parameter and longpulse plasma operation in EAST.The experiences also would be constructive for us to promote the understanding of relevant physics and basis towards the ITER-like condition.