检索结果-内蒙古大学图书馆

Scheduling of Low-Latency Medical Services in Healthcare Cloud with Deep Reinforcement Learning

Tsinghua science and technology 2025年第1期30卷 100-111页

作者： Hongfei Du Ming Liu Nianbo Liu Deying Li Wenzhong Li Lifeng Xu School of Computer Science and Engineering University of Electronic Science and Technology of ChinaChengdu 610054China School of Information Renmin University of ChinaBeijing 100872China Department of Computer Science and Technology Nanjing UniversityNanjing 210093China People’s Hospital of Quzhou City Wenzhou Medical UniversityQuzhou 324000China

In the current landscape of online data services,data transmission and cloud computing are often controlled separately by Internet Service Providers(ISPs)and cloud providers,resulting in significant cooperation challenges and suboptimal global data service *** this study,we propose an end-to-end scheduling method aimed at supporting low-latency and computation-intensive medical services within local wireless networks and healthcare *** approach serves as a practical paradigm for achieving low-latency data services in local private cloud *** meet the low-latency requirement while minimizing communication and computation resource usage,we leverage Deep Reinforcement Learning(DRL)algorithms to learn a policy for automatically regulating the transmission rate of medical services and the computation speed of cloud ***,we utilize a two-stage tandem queue to address this problem *** experiments are conducted to validate the effectiveness for our proposed method under various arrival rates of medical services.

关键词： medical service tandem queue cloud computing Deep Reinforcement Learning(DRL) resource allocation

来源：评论

学校读者我要写书评

暂无评论

Early identification of high-risk patients admitted to emergency departments using vital signs and machine learning

引用

World Journal of Emergency Medicine 2025年第2期16卷 113-120页

作者： Qingyuan Liu Yixin Zhang Jian Sun Kaipeng Wang Yueguo Wang Yulan Wang Cailing Ren Yan Wang Jiashan Zhu Shusheng Zhou Mengping Zhang Yinglei Lai Kui Jin School of Mathematics and Physics Anhui Jianzhu UniversityHefei 230601China School of Mathematical Sciences University of Science and Technology of ChinaHefei 230026China Department of Emergency Medicine the First Affiliated Hospital of University of Science and Technology of ChinaDivision of Life Sciences and MedicineUniversity of Science and Technology of ChinaHefei 230001China School of Mathematics and Statistics Nanjing University of Science and TechnologyNanjing 210094China

BACKGROUND:Rapid and accurate identification of high-risk patients in the emergency departments(EDs)is crucial for optimizing resource allocation and improving patient *** study aimed to develop an early prediction model for identifying high-risk patients in EDs using initial vital sign ***:This retrospective cohort study analyzed initial vital signs from the Chinese Emergency Triage,Assessment,and Treatment(CETAT)database,which was collected between January 1^(st),2020,and June 25^(th),*** primary outcome was the identification of high-risk patients needing immediate *** machine learning methods,including a deep-learningbased multilayer perceptron(MLP)classifier were *** performance was assessed using the area under the receiver operating characteristic curve(AUC-ROC).AUC-ROC values were reported for three scenarios:a default case,a scenario requiring sensitivity greater than 0.8(Scenario I),and a scenario requiring specificity greater than 0.8(Scenario II).SHAP values were calculated to determine the importance of each predictor within the MLP ***:A total of 38,797 patients were analyzed,of whom 18.2%were identified as *** analysis of the predictive models for high-risk patients showed AUC-ROC values ranging from 0.717 to 0.738,with the MLP model outperforming logistic regression(LR),Gaussian Naive Bayes(GNB),and the National Early Warning Score(NEWS).SHAP value analysis identified coma state,peripheral capillary oxygen saturation(SpO_(2)),and systolic blood pressure as the top three predictive factors in the MLP model,with coma state exerting the most ***:Compared with other methods,the MLP model with initial vital signs demonstrated optimal prediction accuracy,highlighting its potential to enhance clinical decision-making in triage in the EDs.

关键词： Machine learning Triage Emergency medicine Decision support systems

来源：评论

学校读者我要写书评

暂无评论

Tailoring precipitate distribution in 2024 aluminum alloy for improving strength and corrosion resistance

引用

Journal of Materials science & technology 2024年第27期194卷 16-27页

作者： Yong Zhang Lantian Zhang Xiang Gao Xulong An Le Zong Zhihong Jia Hao Zhou Yudong Sui Wenwen Sun Jiangsu Key Laboratory for Advanced Metallic Materials School of Materials Science and EngineeringSoutheast UniversityNanjing 211189China Department of Materials Science and Engineering Monash UniversityClaytonVICAustralia School of Materials Science and Engineering Changzhou UniversityChangzhou 213164China Key Laboratory for Light-weight Materials Nanjing Tech UniversityNanjing 210009China Nano and Heterogeneous Materials Center School of Materials Science and EngineeringNanjing University of Science and TechnologyNanjing 210094China School of Materials Science and Engineering Kunming University of Science and TechnologyKunming 650093China

For the traditional peak-aged (PA) AA2024 alloy, the formation of large S-phase precipitates within the grains, wide precipitate-free zones (PFZs) near the grain boundaries (GBs), and continuous distribution of grain boundary precipitates (GBPs) can be observed. As a result, the PA alloy exhibits relatively high strength but poor corrosion resistance. However, with the application of cyclic plasticity treatment, high-density 1–2 nm clusters form within the matrix, and no PFZs form near GBs. In this study, this treatment yields the optimal balance between strength–elongation characteristics and corrosion resistance. By combining cyclic plasticity and ageing heat treatment with different heating rates, the nanoscale clusters play a crucial role as heterogeneous nucleation sites, resulting in the formation of finer and higher number density of S precipitates within the matrix. Additionally, the presence of these clusters reduces the formation of GBPs and minimizes the width of PFZs. Consequently, compared to the traditional PA sample, this approach achieves a significantly higher yield strength (increased by 46 %) and ultimate tensile strength (increased by 18 %), along with superior corrosion resistance. Although the influence of ageing heat treatment with different rates on mechanical properties is not significant, it notably affects the formation of GBPs and corrosion resistance. Specifically, a slower heating rate leads to an increase in the spacing between adjacent GBPs, resulting in improved corrosion resistance. In summary, cyclic strengthening, as a novel method for alloy strengthening, when combined with ageing heat treatment, modulates the distribution of S precipitates within the matrix and GBs. This optimization maximizes the effects of precipitation strengthening and breaks the inverse relationship between strength and corrosion resistance.

关键词： Aluminum alloys Mechanical properties Intergranular corrosion Cyclic plasticity Precipitate-free zone Grain boundary precipitates

来源：评论

学校读者我要写书评

暂无评论

Inverse Design and Experimental Verification of Metamaterials for Thermal Illusion Using Genetic Algorithms

引用

Chinese Physics Letters 2023年第10期40卷 73-77页

作者：何宗堽袁坤熊国欢王健 College of Physical Science and Technology Yangzhou UniversityYangzhou 225002China School of Physical Science and Technology Nanjing Normal UniversityNanjing 210046China

Thermal metamaterials offer a promising avenue for creating artificial materials with unconventional physical properties,such as thermal cloak,concentrator,rotator,and ***,designs and fabrication of thermal metamaterials are of challenge due to the limitations of existing methods on anisotropic material *** propose an evolutionary framework for designing thermal metamaterials using genetic algorithm *** approach encodes unit cells with different thermal conductivities and performs global optimization using the evolution-inspired *** further fabricate the thermal functional cells using 3D printing and verify their thermal illusion functionality *** study introduces a new design paradigm for advanced thermal metamaterials that can manipulate heat flows robustly and realize functional thermal metadevices without anisotropic thermal *** approach can be easily applied to fabrications in various fields such as thermal management and thermal sensing.

关键词： anisotropic thermal Inverse

来源：评论

学校读者我要写书评

暂无评论

Rare earth lanthanum pinning effect for corrosion resistance ultraefficient microwave absorption FeCo@rGO composites

引用

Journal of Materials science & technology 2024年第10期177卷 181-190页

作者： Junru Yao Jintang Zhou Lu Lu Feng Yang Zhengjun Yao Bo Ouyang Erjun Kan Yuxin Zuo Renchao Che Fan Wu College of Materials Science and Technology Nanjing University of Aeronautics and AstronauticsNanjing 211100China Ministry of Industry and Information Technology Key Laboratory of Material Preparation and Protection for Harsh Environment(Nanjing University of Aeronautics and Astronautics)Nanjing 211100China MIIT Key Laboratory of Semiconductor Microstructure and Quantum Sensing School of ScienceNanjing University of Science and TechnologyNanjing 210094China School of Chemistry and Chemical Engineering Anhui UniversityHefei 230601China Laboratory of Advanced Materials Shanghai Key Lab of Molecular Catalysis and Innovative MaterialsDepartment of Materials ScienceFudan UniversityShanghai 200438China School of Mechanical Engineering Nanjing University of Science and TechnologyNanjing 210094China

Magnetic metal absorbers exhibit remarkable microwave absorption ***,their practical application is severely limited due to their susceptibility to corrosion,particularly in marine *** address this challenge,we propose a novel approach involving the modification and control of FeCo/rGO microwave absorbers using rare earth lanthanum(La).This strategy aims to achieve both high-performance microwave absorption and enhanced resistance to marine *** this study,we employ a La_(2)O_(3) modifying control strategy to refine the FeCo magnetic particles and coat them with CoFe_(2)O_(4) on the surface,leveraging the pinning effect of in situ generated La_(2)O_(3) .This process enhances the interface polarization of the absorbers,thereby improving their electromagnetic performance and ma-rine corrosion ***,the La_(2)O_(3) modified FeCo@rGO composites exhibit broadband ab-sorption,covering a wide frequency range of 6.11 GHz at 1.55 ***,the electromagnetic proper-ties of the La_(2)O_(3) modified FeCo@rGO absorbers remain stable even after prolonged exposure to a 3.5 wt% NaCI solution,simulating marine conditions,for at least 15 ***,we perform first-principle calculations on FeCo and FeCo-O to validate the corrosion resistance of the La_(2)O_(3) modified FeCo@rGO composites at the atomic *** comprehensive investigation explores the control of rare earth lan-thanum modification on the size of magnetic metal particles,enabling efficient electromagnetic wave absorption and marine corrosion *** results of this study provide a novel and facile strategy for the control of microwave absorbers,offering promising prospects for future research and development in this field.

关键词： Rare earth lanthanum FeCo@rGO Microwave absorption Marine corrosion resistance

来源：评论

学校读者我要写书评

暂无评论

Diabetic Retinopathy Image Segmentation Method Based on Fusion DenseNet and U-Net Network

引用

Engineering Letters 2025年第2期33卷 418-428页

作者： Zhang, Yaxuan Xu, Yang Zhai, Jinlong University of Science and Technology LiaoNing Anshan114051 China School of Computer Science and Software Engineering University of Science and Technology LiaoNing Anshan114051 China

Diabetic Retinopathy (DR) is a common and significant complication in patients with diabetes, and severely affecting their quality of life. Image segmentation plays a crucial role in the early diagnosis and treatment of DR. However, traditional methods are limited in terms of segmentation accuracy and generalization capability. This paper proposes a novel image segmentation method for diabetic retinopathy based on the integration of DenseNet and U-Net networks. Firstly, DenseNet is utilized to replace the encoder part of U-Net. The dense connection mechanism enhances the efficiency of feature propagation and improves the feature extraction capability of the encoder. Secondly, we introduce Omni-Dimensional Dynamic Convolution (ODConv) to replace traditional convolutions. ODConv is used to handle diverse input features effectively. The model's adaptability and segmentation accuracy for various samples is improved. Finally, we integrate the Convolutional Block Attention Module (CBAM) into the decoder. By leveraging channel and spatial attention, the ability to capture key features is enhanced, and the segmentation accuracy and boundary recognition capabilities are improved. To validate the effectiveness and feasibility of the model. This paper conducts experiments on the DDR dataset and the IDRID dataset. The experimental results show that the proposed method improved the mDice score by 0.53% compared to traditional methods. It also increased the mIoU score by 1.12% over traditional methods. This indicates that the proposed method has better performance in segmenting diabetic retinopathy images. It also demonstrates better generalization ability. © 2025, International Association of Engineers. All rights reserved.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Microfluidic preparation of surfactant-free ultrafine DAAF with tunable particle size for insensitive initiator explosives

引用

Defence technology（防务技术） 2024年第9期39卷 42-52页

作者： Bo Yang Rui Li Wei Cao Si-min He Jincan Zhu Qi Wu Heng Ding Jin Chen Weimiao Wang Zhiqiang Qiao Xiaodong Li Guangcheng Yang Institute of Chemical Materials China Academy of Engineering PhysicsMianyang 621900China Southwest University of Science and Technology Mianyang 621010China School of Chemical Engineering Nanjing University of Science and TechnologyNanjing 210094China

High purity and ultrafine DAAF(u-DAAF)is an emerging insensitive charge in *** there are many ways to obtain u-DAAF,developing a preparation method with stable operation,accurate control,good quality consistency,equipment miniaturization,and minimum manpower is an inevitable requirement to adapt to the current social technology development *** reported is the microfluidic preparation of u-DAAF with tunable particle size by a passive swirling *** the guidance of recrystallization growth kinetics and mixing behavior of fluids in the swirling microreactor,the key parameters(liquid flow rate,explosive concentration and crystallization temperature)were screened and optimized through screening *** the condition that no surfactant is added and only experimental parameters are controlled,the particle size of recrystallized DAAF can be adjusted from 98 nm to 785 nm,and the corresponding specific surface area is 8.45 m^(2)·g^(-1)to 1.33 m^(2)·g^(-1).In addition,the preparation method has good batch stability,high yield(90.8%-92.6%)and high purity(99.0%-99.4%),indicating a high practical application *** explosion derived flyer initiation tests demonstrate that the u-DAAF shows an initiation sensitivity much lower than that of the raw DAAF,and comparable to that of the refined DAAF by conventional spraying crystallization *** study provides an efficient method to fabricate u-DAAF with narrow particle size distribution and high reproducibility as well as a theoretical reference for fabrication of other ultrafine explosives.

关键词： Microfluidic preparation Screening crystallization conditions Narrow particle size distribution Low initiation sensitivity Ultrafine DAAF

来源：评论

学校读者我要写书评

暂无评论

Federated Learning Based on Data Divergence and Differential Privacy in Financial Risk Control Research

引用

computers, Materials & Continua 2023年第4期75卷 863-878页

作者： Mao Yuxin Wang Honglin School of Artificial Intelligence/School of Future Technology Nanjing University of Information Science and TechnologyNanjing210044China

In the financial sector, data are highly confidential and sensitive,and ensuring data privacy is critical. Sample fusion is the basis of horizontalfederation learning, but it is suitable only for scenarios where customershave the same format but different targets, namely for scenarios with strongfeature overlapping and weak user overlapping. To solve this limitation, thispaper proposes a federated learning-based model with local data sharing anddifferential privacy. The indexing mechanism of differential privacy is used toobtain different degrees of privacy budgets, which are applied to the gradientaccording to the contribution degree to ensure privacy without affectingaccuracy. In addition, data sharing is performed to improve the utility ofthe global model. Further, the distributed prediction model is used to predictcustomers’ loan propensity on the premise of protecting user privacy. Usingan aggregation mechanism based on federated learning can help to train themodel on distributed data without exposing local data. The proposed methodis verified by experiments, and experimental results show that for non-iiddata, the proposed method can effectively improve data accuracy and reducethe impact of sample tilt. The proposed method can be extended to edgecomputing, blockchain, and the Industrial Internet of Things (IIoT) *** theoretical analysis and experimental results show that the proposedmethod can ensure the privacy and accuracy of the federated learning processand can also improve the model utility for non-iid data by 7% compared tothe federated averaging method (FedAvg).

关键词： Data privacy federated learning machine learning data difference

来源：评论

学校读者我要写书评

暂无评论

Efficient stochastic human motion prediction via consistency model 4

Efficient stochastic human motion prediction via consistency...

引用

4th International Conference on computer Vision, Application, and Algorithm, CVAA 2024

作者： Pan, Renhao School of Computer Science and Engineering Nanjing University of Science and Technology Jiangsu Nanjing210094 China

ISBN: (数字)9781510687622

ISBN: (纸本)9781510687615

Diffusion models have recently proven successful in stochastic human motion prediction. Despite their excellent generative performance, they are difficult to predict in real-time because the multi-step sampling mechanism involves tens or hundreds of iterations of repetitive function evaluation. To adress this issue, we introduce the Motion Consistency Model (MotionCM) to alleviate the computational and time consumption of the iterative inference process. It applies a diffusion pipelline to the low-frequency domain processed by discrete cosine transformation(DCT), alleviating the computational burden for each function evaluation. The predictive efficiency of the motion diffusion model in stochastic human motion is further improved by employing a one-step (or several-step) inference via maitaining consistency of the outputs over the trajectory of PF-ODE. Extensive experiments on two benchmark datasets show that the proposed model achieves state-of-the-art performance at less than 10% of the time cost. © 2025 SPIE.

关键词： Cosine transforms

来源：评论

学校读者我要写书评

暂无评论

RASTMTrack: Robust and Adaptive Space-Time Memory Networks for Visual Tracking 14

RASTMTrack: Robust and Adaptive Space-Time Memory Networks f...

引用

14th IEEE International Conference on Cyber technology in Automation, Control, and Intelligent Systems, CYBER 2024

作者： Zheng, Yu Liu, Yong Che, Xun School of Computer Science and Engineering Nanjing University of Science and Technology Jiangsu Xiaolingwei Nanjing210094 China School of Computer Science and Engineering Nanjing University of Science and Technology Xiaolingwei Jiangsu Nanjing210094 China

ISBN: (纸本)9798331506056

These trackers based on the space-time memory network locate the target object in the search image employing contextual information from multiple memory frames and their corresponding foreground-background features. It is conceivable that these trackers are susceptible to the memory frame quality as well as the accuracy of the corresponding foreground labels. In the previous work (STMTrack), a fixed and experienced extraction strategy for memory frames was adopted, which hinders the improvement of performance and generalization. To address these limitations, we propose a dynamic and adaptive extraction strategy for memory frames to ensure that the most representative historical frames are selected into the set of memory frames to improve the accuracy of localization and reduce failures due to error accumulation. Specifically, we propose an extraction network to evaluate historical frames, where historical frames with the highest score are labeled as the memory frame and conversely dropped. Qualitative and quantitative analyses were implemented on multiple datasets (LaSOT and GOT-10K), and our method achieves significant performance gains (LaSOT: 60.6→61.4, GOT-10K: 64.2→65.2) over STMTrack especially for challenging scenarios (e.g., occlusion, vanishing, and reappearance, etc.). while bringing only a negligible inference speed degradation (Fps: 37 → 34), otherwise, the proposed method obtains competitive results compared to other state-of-the-art (SOTA) methods (CNN-based). © 2024 IEEE.

关键词： Space time adaptive processing

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：