检索结果-南通市图书馆

International Journal of Ophthalmology(English edition) 2024年第10期17卷 1763-1771页

作者： chen Yu Lin hung ju chen Yi Kit Chan Wei Ping Hsia Yu Len Huang Chia Jen Chang Department of Ophthalmology Taichung Veterans General HospitalTaichung 407TaiwanChina Department of Computer Science Tunghai UniversityTaichung 407TaiwanChina

AIM:To develop an automated model for subfoveal choroidal thickness(SFCT)detection in optical coherence tomography(OCT)images,addressing manual fovea location and choroidal contour ***:Two procedures were proposed:defining the fovea and segmenting the *** localization from B-scan OCT image sequence with three-dimensional reconstruction(LocBscan-3D)predicted fovea location using central foveal depression features,and fovea localization from two-dimensional en-face OCT(LocEN-2D)used a mask region-based convolutional neural network(Mask R-CNN)model for optic disc detection,and determined the fovea location based on optic disc relative *** segmentation also employed Mask ***:For 53 eyes in 28 healthy subjects,LocBscan-3D’s mean difference between manual and predicted fovea locations was 170.0μm,LocEN-2D yielded 675.9μ***-2D performed better in non-high myopia group(P=0.02).SFCT measurements from Mask R-CNN aligned with manual ***:Our models accurately predict SFCT in OCT ***-3D excels in precise fovea localization even with high ***-2D shows high detection rates but lower accuracy especially in the high myopia *** both models offers a robust SFCT assessment approach,promising efficiency and accuracy for large-scale studies and clinical use.

关键词： subfoveal choroidal thickness optical coherence tomography automatic foveal detection automatic choroid segmentation

On convergence of covariance matrix of empirical Bayes hyper-parameter estimator

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Control Theory and Technology 2024年第2期22卷 149-162页

作者： Yue ju Biqiang Mu Tianshi chen School of Data Science and Shenzhen Research Institute of Big Data The Chinese University of Hong KongShenzhenShenzhen518172GuangdongChina Division of Decision and Control Systems School of Electrical Engineering and Computer ScienceKTH Royal Institute of Technology10044StockholmSweden Key Laboratory of Systems and Control Institute of Systems ScienceAcademy of Mathematics and System ScienceChinese Academy of SciencesBeijing100190China

Regularized system identification has become the research frontier of system identification in the past *** related core subject is to study the convergence properties of various hyper-parameter estimators as the sample size goes to *** this paper,we consider one commonly used hyper-parameter estimator,the empirical Bayes(EB).Its convergence in distribution has been studied,and the explicit expression of the covariance matrix of its limiting distribution has been ***,what we are truly interested in are factors contained in the covariance matrix of the EB hyper-parameter estimator,and then,the convergence of its covariance matrix to that of its limiting distribution is *** general,the convergence in distribution of a sequence of random variables does not necessarily guarantee the convergence of its covariance ***,the derivation of such convergence is a necessary complement to our theoretical analysis about factors that influence the convergence properties of the EB hyper-parameter *** this paper,we consider the regularized finite impulse response(FIR)model estimation with deterministic inputs,and show that the covariance matrix of the EB hyper-parameter estimator converges to that of its limiting ***,we run numerical simulations to demonstrate the efficacy of ourtheoretical results.

关键词： Regularized system identification Hyper-parameter estimator Empirical Bayes Convergence of covariance matrix

Durable Ru Nanocrystal with HfO_(2)Modification for Acidic Overall Water Splitting

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Nano-Micro Letters 2024年第9期16卷 462-477页

作者： Xiangkai Kong Jie Xu Zhicheng ju Changle chen School of Materials and Physics China University of Mining and TechnologyXuzhou 221116JiangsuPeople’s Republic of China Anhui Province Key Laboratory of Pollutant Sensitive Materials and Environmental Remediation Huaibei Normal UniversityHuaibei 235000AnhuiPeople’s Republic of China School of Chemistry and Materials Science University of Science and Technology of ChinaHefei 230026AnhuiPeople’s Republic of China School of Chemistry and Chemical Engineering Nanjing UniversityNanjing 210023JiangsuPeople’s Republic of China

Durable and efficient bi-functional catalyst,that is capable of both oxygen evolution reaction and hydrogen evolution reaction under acidic condition,are highly desired for the commercialization of proton exchange membrane water ***,we report a robust L-Ru/HfO_(2)heterostructure constructed via confining crystalline Ru nanodomains by HfO_(2)*** assembled with a proton exchange membrane,the bi-functional L-Ru/HfO_(2)catalyst-based electrolyzer presents a voltage of 1.57 and 1.67 V to reach 100 and 300 mA cm^(-2)current density,prevailing most of previously reported Ru-based materials as well as commercial Pt/C||RuO_(2)*** is revealed that the synergistic effect of HfO_(2)modification and small crystalline domain formation significantly alleviates the over-oxidation of *** importantly,this synergistic effect facilitates a dual-site oxide path during the oxygen evolution procedure via optimization of the binding configurations of oxygenated *** a result,the Ru active sites maintain the metallic state along with reduced energy barrier for the rate-determining step(^(*)O→^(*)OOH).Both of water adsorption and dissociation(Volmer step)are strengthened,while a moderate hydrogen binding is achieved to accelerate the hydrogen desorption procedure(Tafel step).Consequently,the activity and stability of acidic overall water splitting are simultaneously enhanced.

关键词： Ruthenium Hafnium dioxide Oxygen evolution catalysis Anti-oxidation

维普期刊数据库

Identifying Brand Consistency by Product Differentiation Using CNN

在线全文

学校读者我要写书评

暂无评论

Computer Modeling in Engineering & Sciences 2024年第7期140卷 685-709页

作者： hung-Hsiang Wang Chih-Ping chen Department of Industrial Design National Taipei University of TechnologyTaipeiTaiwan Department of Product Design Ming Chuan UniversityTaoyuanTaiwan

This paper presents a new method of using a convolutional neural network(CNN)in machine learning to identify brand consistency by product appearance *** Experiment 1,we collected fifty mouse devices from the past thirty-five years from a renowned company to build a dataset consisting of product pictures with pre-defined design features of their appearance and *** show that it is a challenge to distinguish periods for the subtle evolution of themouse devices with such traditionalmethods as time series analysis and principal component analysis(PCA).In Experiment 2,we applied deep learning to predict the extent to which the product appearance variation ofmouse devices of various *** investigation collected 6,042 images ofmouse devices and divided theminto the Early Stage and the Late *** show the highest accuracy of 81.4%with the CNNmodel,and the evaluation score of brand style consistency is 0.36,implying that the brand consistency score converted by the CNN accuracy rate is not always perfect in the real *** relationship between product appearance variation,brand style consistency,and evaluation score is beneficial for predicting new product styles and future product style *** addition,the CNN heat maps highlight the critical areas of design features of different styles,providing alternative clues related to the blurred *** study provides insights into practical problems for designers,manufacturers,and marketers in product *** not only contributes to the scientific understanding of design development,but also provides industry professionals with practical tools and methods to improve the design process and maintain brand *** can use these techniques to find features that influence brand ***,capture these features as innovative design elements and maintain core brand values.

关键词： Machine learning product differentiation brand consistency principal component analysis convolutional neural network computer mouse

Paraffin–CaCl_(2)·6H_(2)O dosage effects on the strength and heat transfer characteristics of cemented tailings backfill

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

International Journal of Minerals,Metallurgy and Materials 2024年第1期31卷 60-70页

作者： Hai Li Aibing Jin Shuaijun chen Yiqing Zhao You ju Key Laboratory of Ministry of Education for Efficient Mining and Safety of Metal Mine University of Science and Technology BeijingBeijing 100083China School of Civil and Resources Engineering University of Science and Technology BeijingBeijing 100083China

The challenge of high temperatures in deep mining remains harmful to the health of workers and their production efficiency The addition of phase change materials (PCMs) to filling slurry and the use of the cold storage function of these materials to reduce downhole temperatures is an effective approach to alleviate the aforementioned ***–CaCl_(2)·6H_(2)O composite PCM was prepared in the *** composition,phase change latent heat,thermal conductivity,and cemented tailing backfill (CTB) compressive strength of the new material were *** heat transfer characteristics and endothermic effect of the PCM were simulated using Fluent *** results showed the following:(1) The new paraffin–CaCl_(2)·6H_(2)O composite PCM improved the thermal conductivity of native paraffin while avoiding the water solubility of CaCl_(2)·6H_(2)O.(2) The calculation formula of the thermal conductivity of CaCl_(2)·6H_(2)O combined with paraffin was deduced,and the reasons were explained in principle.(3) The“enthalpy–mass scale model”was applied to calculate the phase change latent heat of nonreactive composite PCMs.(4)The addition of the paraffin–CaCl_(2)·6H_(2)O composite PCM reduced the CTB strength but increased its heat absorption *** research can give a theoretical foundation for the use of heat storage backfill in green mines.

关键词： paraffin–CaCl_(2)·6H_(2)O heat transfer simulation heat calculation phase change material-based backfill latent heat of formula

维普期刊数据库

Towards the universal defense for query-based audio adversarial attacks on speech recognition system

在线全文

学校读者我要写书评

暂无评论

Cybersecurity 2024年第1期7卷 53-70页

作者： Feng Guo Zheng Sun Yuxuan chen Lei ju School of Cyber Science and Technology Shandong UniversityQingdaoChina Quancheng Laboratory QCLJinanChina

Recently,studies show that deep learning-based automatic speech recognition(ASR)systems are vulnerable to adversarial examples(AEs),which add a small amount of noise to the original audio *** AE attacks pose new challenges to deep learning security and have raised significant concerns about deploying ASR systems and *** existing defense methods are either limited in application or only defend on results,but not on *** this work,we propose a novel method to infer the adversary intent and discover audio adversarial examples based on the AEs generation *** insight of this method is based on the observation:many existing audio AE attacks utilize query-based methods,which means the adversary must send continuous and similar queries to target ASR models during the audio AE generation *** by this observation,We propose a memory mechanism by adopting audio fingerprint technology to analyze the similarity of the current query with a certain length of memory ***,we can identify when a sequence of queries appears to be suspectable to generate audio *** extensive evaluation on four state-of-the-art audio AE attacks,we demonstrate that on average our defense identify the adversary’s intent with over 90%*** careful regard for robustness evaluations,we also analyze our proposed defense and its strength to withstand two adaptive ***,our scheme is available out-of-the-box and directly compatible with any ensemble of ASR defense models to uncover audio AE attacks effectively without model retraining.

关键词： Adversarial attacks Defense Memory mechanism Query-based

Heterogeneous decentralised machine unlearning with seed model distillation

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

CAAI Transactions on Intelligence Technology 2024年第3期9卷 608-619页

作者： Guanhua Ye Tong chen Quoc Viet hung Nguyen Hongzhi Yin School of Information Technology and Electrical Engineering The University of QueenslandSt LuciaQueenslandAustralia School of Information and Communication Technology Griffith UniversityBrisbaneQueenslandAustralia

As some recent information security legislation endowed users with unconditional rights to be forgotten by any trained machine learning model,personalised IoT service pro-viders have to put unlearning functionality into their *** most straight-forward method to unlearn users'contribution is to retrain the model from the initial state,which is not realistic in high throughput applications with frequent unlearning *** some machine unlearning frameworks have been proposed to speed up the retraining process,they fail to match decentralised learning scenarios.A decentralised unlearning framework called heterogeneous decentralised unlearning framework with seed(HDUS)is designed,which uses distilled seed models to construct erasable en-sembles for all ***,the framework is compatible with heterogeneous on-device models,representing stronger scalability in real-world *** experiments on three real-world datasets show that our HDUS achieves state-of-the-art performance.

关键词： data mining data privacy machine learning

Evidence-Based Dampness-Heat ZHENG(Syndrome) in Cancer: Current Progress toward Establishing Relevant Animal Model with Pancreatic Tumor

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Chinese Journal of Integrative Medicine 2024年第1期30卷 85-95页

作者： JIAO ju-ying chenG Chien-shan CAO Zhang-qi chen Lian-yu chen Zhen Department of Integrative Oncology Fudan University Shanghai Cancer CenterShanghai(200032)China Department of Oncology Shanghai Medical CollegeFudan UniversityShanghai(200032)China

Cancer is one of the deadliest diseases affecting the health of human beings. With limited therapeutic options available, complementary and alternative medicine has been widely adopted in cancer management and is increasingly becoming accepted by both patients and healthcare workers alike. Chinese medicine characterized by its unique diagnostic and treatment system is the most widely applied complementary and alternative medicine. It emphasizes symptoms and ZHENG(syndrome)-based treatment combined with contemporary disease diagnosis and further stratifies patients into individualized medicine subgroups. As a representative cancer with the highest degree of malignancy, pancreatic cancer is traditionally classified into the "amassment and accumulation". Emerging perspectives define the core pathogenesis of pancreatic cancer as "dampness-heat" and the respective treatment "clearing heat and resolving dampness" has been demonstrated to prolong survival in pancreatic cancer patients, as has been observed in many other cancers. This clinical advantage encourages an exploration of the essence of dampness-heat ZHENG(DHZ) in cancer and investigation into underlying mechanisms of action of herbal formulations against dampness-heat. However, at present, there is a lack of understanding of the molecular characteristics of DHZ in cancer and no standardized and widely accepted animal model to study this core syndrome in vivo. The shortage of animal models limits the ability to uncover the antitumor mechanisms of herbal medicines and to assess the safety profile of the natural products derived from them. This review summarizes the current research on DHZ in cancer in terms of the clinical aspects, molecular landscape, and animal models. This study aims to provide comprehensive insight that can be used for the establishment of a future standardized ZHENG-based cancer animal model.

关键词： cancer Chinese medicine syndrome ZHENG dampness-heat syndrome animal model pancreatic tumor evaluation methods

维普期刊数据库

Sedimentary paleoenvironment and its control on organic matter enrichment in the Mesoproterozoic hydrocarbon source rocks in the Ordos Basin, southern margin of the North China Craton

在线全文

学校读者我要写书评

暂无评论

Petroleum Science 2024年第4期21卷 2257-2272页

作者： Zhi-chen Wu ju-Ye Shi Tai-Liang Fan Ming Jiang School of Energy Resources China University of Geosciences (Beijing) Key Laboratory of Marine Reservoir Evolution and Hydrocarbon Enrichment Mechanism Ministry of Education China University of Geosciences (Beijing)

The black shale of the Mesoproterozoic Cuizhuang Formation in the Changcheng System in Yongji city,North China Craton, is a potential source rock. Understanding the organic matter enrichment mechanism is crucial for evaluating source rock resources and understanding oil and gas accumulation *** this study, we evaluated the sedimentary paleoenvironment and organic matter enrichment mechanisms of shale using thin section observations, mineral composition analysis, organic geochemistry, and elemental geochemistry. We found significant differences in the sedimentary paleoenvironment and organic matter enrichment mechanisms between the lower Cuizhuang Formation and the Beidajian Formation shale. The Cuizhuang Formation was deposited in a late-stage, restricted basin environment during the rift phase, and elemental and geochemical indicators showed that the Cuizhuang Formation was in a suboxic-anoxic water environment, that was influenced by a warm and humid paleoclimate and submarine hydrothermal activities, which promoted the accumulation of organic matter. However, the enrichment of organic matter in the Cuizhuang Formation was mainly controlled by redox *** formation of suboxic-anoxic water environments may be closely related to the warm and humid paleoclimate and submarine hydrothermal activities. Warm conditions promote continental weathering and increase marine productivity, thereby consuming oxygen in the bottom water. Moreover, acidic hydrothermal activity also helps to establish an anoxic environment. Our results reveal the effects controlling various coupled mechanisms dominated by redox conditions, which may explain the development of source rocks in the Cuizhuang Formation.

关键词： Cuizhuang formation Paleoproterozoic source rocks Marine productivity Paleoenvironments Submarine hydrothermal activity Organic matter enrichment

A smart hybrid memory scheduling approach using neural models

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2024年第4期67卷 333-334页

作者： Yanjie ZHEN Huijun ZHANG Yongheng DENG Weining chen Wei GAO ju REN Yu chen Department of Computer Science and Technology Tsinghua University China Huaneng Clean Energy Research Institute

Traditional dynamic random access memory is limited by its low capacity and high cost because of the increasing demand for the memory capacity of applications. Hybrid memory architectures, including new memory hardwar... 详细信息

关键词： Cost reduction