检索结果-内蒙古大学图书馆

Machine learning-based classification of medication adherence among patients with noncommunicable diseases

Informatics in Medicine Unlocked 2025年 52卷

作者： Kanyongo, Wellington Ezugwu, Absalom E. Moyo, Tsitsi Fonou Dombeu, Jean Vincent Department of Computer Science Faculty of Science Engineering Bindura University of Science Education Bindura Zimbabwe Unit for Data Science and Computing North-West University 11 Hoffman Street Potchefstroom 2520 South Africa Cimas Medical Aid Society Borrowdale Office Park Borrowdale Road P. O Box 1243 Harare Zimbabwe School of Mathematics Statistics and Computer Science University of KwaZulu-Natal Pietermaritzburg Campus Pietermaritzburg 3201 South Africa

Non-adherence to medication among individuals with non-communicable diseases (NCDs) leads to increased morbidity, mortality, and healthcare costs. The integration of electronic drug prescription and dispensation systems enables comprehensive analysis of medication adherence (MA). Patient-level and medical claims data for 8141 diabetic and hypertensive patients in Harare, Zimbabwe, were analysed. Non-adherence was defined as medication refills falling below 75 % of the intended 12 monthly claims, while adherence required at least 75 % of the refills. Classification employed multiple machine learning algorithms, including SVM, KNN, DT, Naïve Bayes, DNN, LR, and RF in Python 3.11.3. Significant variables for MA were identified through the Random Forest (RF) feature importance mechanism and the information gain technique. These included the annual quantity of medical supplies, annual claim amount, patient age, wellness program subscription, medical aid cover, contribution towards medical aid cover, comorbidity, diagnosis, hospital cover type, complications development, gender, and medical aid scheme. The total units of medical supplies dispensed annually emerged as the most significant predictor of MA. Considering the 8-feature subset, which consistently produced the most robust machine learning models, the classification accuracy of the ML classifiers ranged from 84.9 % to 88.2 %, while the AUC values varied between 0.857 and 0.934. RF, an ensemble learning technique, was the most robust classifier, achieving 88.2 % accuracy, an AUC of 0.935, and superior precision, recall, and F1-score. This model shows promise as a prognostic tool for enhancing MA, aiding in identifying adherence levels among patients. These findings contribute to addressing disparities in medication refilling and adherence rates among patients with NCDs. The ML model holds potential for the development of intelligent MA and intervention applications to improve patient adherence to medication in the

关键词： Chronic disease Decision trees Deep neural network Diabetes Hypertension k-NN Logistic regression Machine learning Medication adherence Naïve bayes Non-communicable diseases Random forest SVM

来源：评论

学校读者我要写书评

暂无评论

Enhanced Framework for MRI Brain Tumor Recognition with Residual Learning and Intuitive Heatmap Visualization

Enhanced Framework for MRI Brain Tumor Recognition with Resi...

引用

Innovative computing, Intelligent Communication and Smart Electrical Systems (ICSES), International Conference on

作者： L. Velmurugan R. K Maheswari K. Janani K. Agalya M. Poornima K.C. Gayathri School of Computing Science and Engineering (SCOPE VIT Bhopal University Sehore Madhya Pradesh India Artificial Intelligence and Data Science M.A.M College of Enginnering and Technology Trichy Tamilnadu India Department of Data science and computer applications Manipal institute of Technology Manipal Academy of Higher Education Manipal India Department of Computer Science Engineering Sri Eshwar College of Engineering Coimbatore India Department of Information Technology OASYS Institute of Technology Trichy Tamil Nadu India Chettinad School of physiotherapy (CSP Chettinad Hospital and Research Institute (CHRI) Chettinad Academy of Research and Education (CARE) Chennai India

ISBN: (数字)9798331543624

ISBN: (纸本)9798331543631

Early detection of brain tumors is crucial for effective treatment, yet it poses a significant challenge due to the rapid nature of the disease's progression. This study explores the application of ResNet, aimed at improving the accuracy of brain tumor diagnosis in magnetic resonance imaging (MRI) scans. Our model was trained on a dataset comprising 253 MRI images, which included 155 confirmed cases of tumors. The residual learning framework of ResNet effectively captures intricate features and mitigates issues such as vanishing gradients. The results indicate that the ResNet model achieved an impressive accuracy rate of 97%, surpassing traditional detection methods. We evaluated the model's performance using metrics like accuracy and F1-score, demonstrating its robustness. To improve the interpretability of our model's predictions, we utilized Grad-CAM (Gradient-weighted Class Activation Mapping), which highlighted critical areas in the MRI scans that influenced the model's decisions. Our findings confirm ResNet’ s effectiveness in medical imaging and stress the importance of model interpretability in healthcare, allowing radiologists to make informed diagnostic choices.

关键词： Deep learning Accuracy Magnetic resonance imaging Brain tumors Medical services Predictive models Brain modeling Robustness Stress Biomedical imaging

来源：评论

学校读者我要写书评

暂无评论

Challenges and Opportunities of DevOps in Cyber-Physical Production Systems Engineering

Challenges and Opportunities of DevOps in Cyber-Physical Pro...

引用

Industrial Cyber-Physical Systems (ICPS)

作者： István Koren Felix Rinker Kristof Meixner Jasminka Matevska Jörg Walter Chair of Process and Data Science RWTH Aachen University Aachen Germany Christian Doppler Laboratory SQI Institute of Information Systems Engineering TU Wien Vienna Austria School of Electrical Engineering and Computer Science City University of Applied Sciences Bremen Germany Distributed Computing and Communication OFFIS - Institute for Information Technology Oldenburg Germany

DevOps is a set of practices that combines software development and operations to enable a continuous software product life cycle to improve the quality of software systems. Although DevOps is considered successful for typical software systems, it has not yet been widelyadopted in the area of - Physlcal Production Systems (CPPSs) that integrate physical components with computer-based control and communication systems. Related to industrial settings, many new challenges arise, like long-term investments, missing flexibility in asset-heavy production environments, and the inherent physicality of hardware. This paper examines the use of the DevOps methodology in the manufacturing domain. It identifies and discusses the unique challenges and describes first solution proposals to overcome those challenges, based on literature and experiences from the industry. The article provides useful guidance to researchers and practitioners on potential pitfalls and exciting opportunities.

关键词： Production systems data collection Software systems Hardware Safety Stakeholders Security

来源：评论

学校读者我要写书评

暂无评论

Shielding Federated Learning: Robust Aggregation with Adaptive Client Selection

arXiv

引用

arXiv 2022年

作者： Wan, Wei Hu, Shengshan Lu, Jianrong Yu Zhang, Leo Jin, Hai He, Yuanyuan School of Cyber Science and Engineering Huazhong University of Science and Technology China School of Computer Science and Technology Huazhong University of Science and Technology China National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Hubei Engineering Research Center on Big Data Security China Cluster and Grid Computing Lab School of Information Technology Deakin University Australia

Federated learning (FL) enables multiple clients to collaboratively train an accurate global model while protecting clients’ data privacy. However, FL is susceptible to Byzantine attacks from malicious participants. Although the problem has gained significant attention, existing defenses have several flaws: the server irrationally chooses malicious clients for aggregation even after they have been detected in previous rounds;the defenses perform ineffectively against sybil attacks or in the heterogeneous data setting. To overcome these issues, we propose MAB-RFL, a new method for robust aggregation in FL. By modelling the client selection as an extended multi-armed bandit (MAB) problem, we propose an adaptive client selection strategy to choose honest clients that are more likely to contribute high-quality updates. We then propose two approaches to identify malicious updates from sybil and non-sybil attacks, based on which rewards for each client selection decision can be accurately evaluated to discourage malicious behaviors. MAB-RFL achieves a satisfying balance between exploration and exploitation on the potential benign clients. Extensive experimental results show that MAB-RFL outperforms existing defenses in three attack scenarios under different percentages of attackers. Copyright © 2022, The Authors. All rights reserved.

关键词： data privacy

来源：评论

学校读者我要写书评

暂无评论

Multi-Agent Reinforcement Learning for Traffic Signal Control through Universal Communication Method

arXiv

引用

arXiv 2022年

作者： Jiang, Qize Qin, Minhao Shi, Shengmin Sun, Weiwei Zheng, Baihua School of Computer Science Fudan University China Shanghai Key Laboratory of Data Science Fudan University China Shanghai Institute of Intelligent Electronics & Systems China School of Computing and Information Systems Singapore Management University Singapore

How to coordinate the communication among intersections effectively in real complex traffic scenarios with multi-intersection is challenging. Existing approaches only enable the communication in a heuristic manner without considering the content/importance of information to be shared. In this paper, we propose a universal communication form UniComm between intersections. UniComm embeds massive observations collected at one agent into crucial predictions of their impact on its neighbors, which improves the communication efficiency and is universal across existing methods. We also propose a concise network UniLight to make full use of communications enabled by UniComm. Experimental results on real datasets demonstrate that UniComm universally improves the performance of existing state-of-the-art methods, and UniLight significantly outperforms existing methods on a wide range of traffic situations. Source codes are available at https://***/zyr17/UniLight. Copyright © 2022, The Authors. All rights reserved.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

data-Driven Active Power Dispatch for Wind Farms

Data-Driven Active Power Dispatch for Wind Farms

引用

data Driven Control and Learning Systems (DDCLS)

作者： Xuguo Jiao Daoyuan Zhang Xin Wang Zhaoxing Ma Zhenyong Zhang Wenfeng Liu School of Information and Control Engineering Qingdao University of Technology Qingdao China State Key Laboratory of Industrial Control Technology College of Control Science and Engineering Zhejiang University Hangzhou China Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center Qilu University of Technology Shandong Academy of Sciences Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China State Key Laboratory of Public Big Data College of Computer Science and Technology Guizhou University Guiyang China School of Civil Engineering Qingdao University of Technology Qingdao China

ISBN: (数字)9798350361674

ISBN: (纸本)9798350361681

Active power dispatch of wind farms plays an important role in power grid scheduling. In this paper, a data-driven active power dispatch strategy for wind farms is proposed, which uses the key point of minimizing the standard deviation of wind turbines' output power series to reduce their fluctuations. First, to build the optimization problem, the historical active power output data of each wind turbine is collected from the Supervisory Control And data Acquisition (SCADA) system. Then, the objective function is developed based on the idea of minimizing the standard deviation of the output power time series. Meanwhile, considering the constraints of wind turbines' output power and active power balance of the grid, the feasible domain of the cost function is constructed. Further, to solve the proposed nonlinear optimization problem with equality and inequality constraints, the trust region interior point method (TRIPM) is employed to explore optimal solution within the feasible domain. Finally, actual data collected from wind farms is used to test effectiveness of the designed approach and experimental results show that our method can achieve better output power smoothing performance than conventional Proportional Dispatch (PD) method.

关键词： Fluctuations Time series analysis SCADA systems Power smoothing Wind farms Linear programming Power grids

来源：评论

学校读者我要写书评

暂无评论

Touchstone benchmark: are we on the right way for evaluating AI algorithms for medical segmentation? 24

Touchstone benchmark: are we on the right way for evaluating...

引用

Proceedings of the 38th International Conference on Neural information Processing Systems

作者： Pedro R. A. S. Bassi Wenxuan Li Yucheng Tang Fabian Isensee Zifu Wang Jieneng Chen Yu-Cheng Chou Saikat Roy Yannick Kirchhoff Maximilian Rokuss Ziyan Huang Jin Ye Junjun He Tassilo Wald Constantin Ulrich Michael Baumgartner Klaus H. Maier-Hein Paul Jaeger Yiwen Ye Yutong Xie Jianpeng Zhang Ziyang Chen Yong Xia Zhaohu Xing Lei Zhu Yousef Sadegheih Afshin Bozorgpour Pratibha Kumari Reza Azad Dorit Merhof Pengcheng Shi Ting Ma Yuxin Du Fan Bai Tiejun Huang Bo Zhao Haonan Wang Xiaomeng Li Hanxue Gu Haoyu Dong Jichen Yang Maciej A. Mazurowski Saumya Gupta Linshan Wu Jiaxin Zhuang Hao Chen Holger Roth Daguang Xu Matthew B. Blaschko Sergio Decherchi Andrea Cavalli Alan L. Yuille Zongwei Zhou Department of Computer Science Johns Hopkins University and Department of Pharmacy and Biotechnology University of Bologna and Center for Biomolecular Nanotechnologies Istituto Italiano di Tecnologia Department of Computer Science Johns Hopkins University NVIDIA Division of Medical Image Computing German Cancer Research Center (DKFZ) and Helmholtz Imaging German Cancer Research Center (DKFZ) ESAT-PSI KU Leuven Division of Medical Image Computing German Cancer Research Center (DKFZ) and Faculty of Mathematics and Computer Science Heidelberg University Division of Medical Image Computing German Cancer Research Center (DKFZ) and Faculty of Mathematics and Computer Science Heidelberg University and HIDSS4Health - Helmholtz Information and Data Science School for Health Shanghai Jiao Tong University Shanghai Artificial Intelligence Laboratory Division of Medical Image Computing German Cancer Research Center (DKFZ) Division of Medical Image Computing German Cancer Research Center (DKFZ) and Pattern Analysis and Learning Group Department of Radiation Oncology Heidelberg University Hospital Helmholtz Imaging German Cancer Research Center (DKFZ) and Interactive Machine Learning Group (IML) DKFZ School of Computer Science and Engineering Northwestern Polytechnical University Australian Institute for Machine Learning The University of Adelaide College of Computer Science and Technology Zhejiang University Hong Kong University of Science and Technology (Guangzhou) Hong Kong University of Science and Technology (Guangzhou) and Hong Kong University of Science and Technology Faculty of Informatics and Data Science University of Regensburg Faculty of Electrical Engineering and Information Technology RWTH Aachen University Faculty of Informatics and Data Science University of Regensburg and Fraunhofer Institute for Digital Medicine MEVIS Electronic & Information Engineering School Harbin Institute of Technology (Shenzhen) Shanghai Jiao Tong University and Beijing Academy of Artificial Intelligence (BAAI) S

ISBN: (纸本)9798331314385

How can we test AI performance? This question seems trivial, but it isn't. Standard benchmarks often have problems such as in-distribution and small-size test sets, oversimplified metrics, unfair comparisons, and short-term outcome pressure. As a consequence, good performance on standard benchmarks does not guarantee success in real-world scenarios. To address these problems, we present Touchstone, a large-scale collaborative segmentation benchmark of 9 types of abdominal organs. This benchmark is based on 5,195 training CT scans from 76 hospitals around the world and 5,903 testing CT scans from 11 additional hospitals. This diverse test set enhances the statistical significance of benchmark results and rigorously evaluates AI algorithms across out-of-distribution scenarios. We invited 14 inventors of 19 AI algorithms to train their algorithms, while our team, as a third party, independently evaluated these algorithms. In addition, we also evaluated pre-existing AI frameworks—which, differing from algorithms, are more flexible and can support different algorithms—including MONAI from NVIDIA, nnU-Net from DKFZ, and numerous other open-source frameworks. We are committed to expanding this benchmark to encourage more innovation of AI algorithms for the medical domain.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Implementing Multi-factor Authentication in Cloud Services for Enhanced Security

Implementing Multi-factor Authentication in Cloud Services f...

引用

Recent Advances in science and Engineering Technology (ICRASET), International Conference on

作者： K. Sashi Rekha V.M. Sivagami R. Usharani T Amutha S. Pushparani Department of Computer Science and Engineering Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Chennai Tamil Nadu India Department of Information Technology Sri Venkateswara College of Engineering Chennai Tamil Nadu India Department of Computational Intelligence School of Computing Faculty of Engineering and Technology SRM Institute of Science and Technology Kattankulathur Chennai India Department of Artificial Intelligence & Data Science CARE College of Engineering Trichy Tamil Nadu India Department of Computer Science and Engineering Meenakshi College of Engineering Chennai Tamil Nadu India

ISBN: (数字)9798350388602

ISBN: (纸本)9798350388619

The cloud services which are now the most common data transmission and endanger organizations’ confidential information, it’s more and more visible that security of any data should be a main priority for companies. MFA, which consists of adopting multi factor approach to authentication and thereby provides multiple types of verification thus ensures effective security, is leveraged to ensure resource access. This paper proposes a full-fledged way of implementing MFA for cloud services in order to strengthen the security given and manage the constatoing risks. The proposed methodology consists of main stages like requirements determining, technology reviewing, risks assessment, technology design, integration with cloud services, users training, pilot deployment and testing, implementation and configuration, monitoring and maintenance as well as constant improvement. Through a systemic approach, the organizations can be equipped with the capability to personalize the MFA solutions according to their security goals, to bring the cloud environments on board, as well as to ensure that the organizations comply with the security standards as set by various governing bodies. This research adds a grain of salt to the development of cloud security good practices through offering a disciplined approach to MFA implementation, which allows partners to harden their cloud assets and protect against growing cyber threats.

关键词： Training Cloud computing Multi-factor authentication Standards organizations Companies Regulation Maintenance Security Protection Monitoring

来源：评论

学校读者我要写书评

暂无评论

Federated Learning for 6G Communications:Challenges,Methods,and Future Directions

引用

China Communications 2020年第9期17卷 105-118页

作者： Yi Liu Xingliang Yuan Zehui Xiong Jiawen Kang Xiaofei Wang Dusit Niyato School of Data Science of Technology Heilongjiang UniversityHarbinChina Faculty of Information Technology Monash UniversityAustralia Alibaba-NTU Joint Research Institute and also School of Computer Science and Engineering NTUSingapore Energy Research Institute Nanyang Technological UniversitySingapore College of Intelligence and Computing Tianjin UniversityTianjinChina School of Computer Science and Engineering NTUSingapore

As the 5G communication networks are being widely deployed worldwide,both industry and academia have started to move beyond 5G and explore 6G *** is generally believed that 6G will be established on ubiquitous Artificial Intelligence(AI)to achieve data-driven Machine Learning(ML)solutions in heterogeneous and massive-scale ***,traditional ML techniques require centralized data collection and processing by a central server,which is becoming a bottleneck of large-scale implementation in daily life due to significantly increasing privacy *** learning,as an emerging distributed AI approach with privacy preservation nature,is particularly attractive for various wireless applications,especially being treated as one of the vital solutions to achieve ubiquitous AI in *** this article,we first introduce the integration of 6G and federated learning and provide potential federated learning applications for *** then describe key technical challenges,the corresponding federated learning methods,and open problems for future research on federated learning in the context of 6G communications.

关键词： 6G communication federated learning security and privacy protection

来源：评论

学校读者我要写书评

暂无评论

FedLRDP: Federated Learning Framework with Local Random Differential Privacy

FedLRDP: Federated Learning Framework with Local Random Diff...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Runtian Zhou Anming Dong Jiguo Yu Qingyan Ding Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center (National Supercomputer Center in Jinan) Qilu University of Technology (Shandong Academy of Sciences) Jinan China School of Computer Science and Technology Qilu University of Technology (Shandong Academy of Sciences) Jinan China Big Data Institute Qilu University of Technology (Shandong Academy of Sciences) Jinan China

ISBN: (数字)9798350359312

ISBN: (纸本)9798350359329

Federated learning (FL) is a distributed machine learning framework enabling multiple clients to collaboratively train a shared ML model without sharing raw data. Despite its aim to safeguard data security and privacy, FL faces risks from advanced adversarial attacks like membership inference attack (MIA), potentially leaking sensitive information. To counter these threats, differential privacy (DP) methods add noise to shared model parameters. However, traditional DP methods struggle with balancing model accuracy, training efficiency, and privacy protection, often compromising one for the other. Additionally, uniform DP mechanisms may not cater to varying client privacy and performance needs. To tackle these challenges, we propose a federated learning framework based on localized random differential privacy (FedLRDP). This approach empowers each client to control noise levels based on their privacy requirements, enhancing model performance while preserving data privacy. We further optimize client-side loss functions to enhance model performance. Theoretical analysis establishes the convergence bounds for FedLRDP, demonstrating its superior convergence performance. Experimental results on MNIST reveal that FedLRDP improves model accuracy by 3.28% compared to traditional DP methods while mitigating MIA with a 72.48% lower attack success rate than FedAvg. Moreover, on EMNIST, FedLRDP boosts model accuracy by 7.31% compared to traditional DP methods, maintaining a 59.4% lower attack success rate than FedAvg. These findings underscore FedLRDP’s efficacy in meeting client privacy needs while enhancing model performance.

关键词： Training Differential privacy Privacy Accuracy Federated learning Noise data models

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：