检索结果-内蒙古大学图书馆

Interdisciplinary Approaches in Technology and Management for Social Innovation (IATMSI), IEEE Conference on

作者： Disha Tuteja Anurag Sinha Deepak Chandra Uprety Dyuti Banerjee Ayushi Dewan Chetna Sharma Department of ECE College Guru Tegh Bahadur Institute of Technology Delhi India Department of Computer Science IGNOU New Delhi India Department of Computer Science and Engineering (CS) School of Computer Science in Emerging Technologies Noida Institute of Engineering and Technology Greater Noida India Department of Artificial Intelligence & Data Science (AI&DS) Koneru Lakshmaiah Education Foundation Green Fields Andhra Pradesh India Rawal Institutions Faridabad India Chitkara University Institute of Engineering and Technology Chitkara University Punjab India

ISBN: (数字)9798331521691

ISBN: (纸本)9798331521707

The COVID-19 pandemic has rapidly spread across the globe, infecting millions of people, with no signs of slowing down. This continuous surge in cases presents a major challenge for healthcare systems worldwide. To combat the exponential rise in infections, medical researchers have primarily relied on RT-PCR testing to distinguish between COVID-19 positive and negative cases. However, this approach faces significant hurdles, particularly in low-income countries where these tests are often too expensive and inaccessible in rural areas. Another critical challenge is the shortage of skilled healthcare professionals, especially doctors and radiologists, whose expertise is in high demand for diagnosing and treating COVID-19 patients. As a result, healthcare resources have been stretched to their limits. To address these issues, innovative solutions are essential, and artificial intelligence (AI) and machine learning (ML) offer promising possibilities. In medical diagnostics, AI, ML, and deep learning (DL) have demonstrated remarkable accuracy. This study explores the application of DL models, specifically ResNet-34 and ResNet-50, for COVID-19 detection using X-ray images. The results indicate an impressive accuracy of 96.4% with ResNet-34. To comprehensively evaluate the performance of these models, the study employs various quantitative metrics, including Matthews Correlation Coefficient (MCC), F1 Score, Precision, and Recall. By contributing to both the effective detection of COVID-19 and addressing the shortage of healthcare professionals, this research provides a hopeful step forward in the ongoing fight against the pandemic.

关键词： COVID-19 Deep learning Measurement Technological innovation Accuracy Pandemics Medical services Public healthcare X-ray imaging Testing

来源：评论

学校读者我要写书评

暂无评论

Enhanced Glaucoma Classification Through Advanced Segmentation by Integrating Cup-to-Disc Ratio and Neuro-Retinal Rim Features

SSRN

引用

SSRN 2025年

作者： Pannu, Rabia Zubair, Muhammad Owais, Muhammad Hassan, Shoaib Umair, Muhammad Usman, Syed Muhammad Al-Bashrawi, Mousa Ahmed Hussain, Irfan Faculty of Information Technology and Computer Science University of Central Punjab Lahore Pakistan Interdisciplinary Research Center for Finance and Digital Economy King Fahd University of Petroleum and Minerals Dhahran Saudi Arabia Department of Mechanical Nuclear Engineering Khalifa University Abu Dhabi United Arab Emirates School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing China Department of Computer Science Bahria School of Engineering and Applied Sciences Bahria University Islamabad Pakistan Department of Information Systems and Operations Management King Fahd University of Petroleum and Minerals Dhahran Saudi Arabia

Glaucoma is a progressive eye condition caused by high intraocular fluid pressure, damaging the optic nerve, leading to gradual, irreversible vision loss, often without noticeable symptoms. Subtle signs such as mild redness in the eyes, slightly blurred vision, and occasional headaches or eye pain may go unnoticed, earning it the nickname "silent thief of sight." Its prevalence is increasing with an aging population, driven by advances in healthcare care and increased life expectancy. Most computer-aided diagnosis (CAD) systems rely on the cup-to-disc ratio (CDR) for the diagnosis of glaucoma. This study introduces a novel approach by integrating CDR with the neuro-retinal rim ratio (NRR), which quantifies rim thickness within the optic disc (OD). NRR improves diagnostic accuracy by recording additional head changes of the optic nerve, such as rim thinning and tissue loss, which were overlooked using CDR alone. We proposed a modified ResUNet architecture for OD and optic cup (OC) segmentation, combining residual learning and U-Net to capture spatial context for semantic segmentation. Its performance was evaluated on the DRISHTI-GS and RIM-ONE datasets. For OC segmentation, the model achieved Dice Coefficient (DC) scores of 0.942 and 0.872 and Intersection over Union (IoU) values of 0.891 and 0.773 for DRISHTI-GS and RIM-ONE, respectively. For OD segmentation, the model achieved DC of 0.972 and 0.950 and IoU values of 0.945 and 0.940 for DRISHTI-GS and RIM-ONE, respectively. CDR and NRR were calculated from segmentation masks and used to train an SVM with a radial basis function, classifying the eyes as healthy or glaucomatous. The model achieved accuracies of 0.969 on DRISHTI-GS and 0.977 on RIM-ONE. © 2025, The Authors. All rights reserved.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

A Parallel Read-Write Circuit With Fast Amplitude-Adaptive Matching Scheme to Memristor Crossbar Array

引用

IEEE Transactions on Circuits and Systems I: Regular Papers 2025年

作者： Jiang, Ningye Jiang, Mingxuan Huang, Haoen Yang, Le Huang, Yi Zeng, Zhigang Huazhong University of Science and Technology School of Artificial Intelligence and Automation Key Laboratory of Image Information Processing and Intelligent Control Ministry of Education of China Wuhan430074 China Wuhan Institute of Technology School of Electrical and Information Engineering Wuhan430205 China University of Massachusetts Amherst Department of Electrical and Computer Engineering AmherstMA01003 United States

Memristor crossbar array (MCA) is a computing-in-memory (CIM) module for computational acceleration. However, conventional read-write (R-W) circuits for MCA rely heavily on external components and have shortcoming in long writing times. To address these issues, we propose a parallel R-W circuit with a fast amplitude-adaptive matching (AAM) scheme. The fast AAM scheme is designed to accelerate time of writing memristors in MCA by adaptively adjusting the writing voltage to match an optimal amplitude. The architecture of parallel R-W circuit are outlined with the implementation of R-W processes. The design principles of parallel R-W unit, fast AAM unit, and associated control logic are described for analyzing their functions within the circuit. By integrating the parallel R-W circuit with MCA, a computing block is leveraged in a neural network for classification tasks. The experimental results demonstrate that the neural network with multiple computing blocks maintains high accuracy across various datasets (over 80%). Furthermore, the proposed AAM scheme achieves an average 54% reduction in writing time compared to that one without AAM scheme. Compared to the pulse-width scheme, the parallel R-W circuit exhibits an average 3.05× speedup in adjusting a 5×5 MCA, and the speedup efficiency is higher as the increases of MCA size. © 2025 IEEE.

关键词： Memristors

来源：评论

学校读者我要写书评

暂无评论

Machine learning strategies for small sample size in materials science

引用

Science China Materials 2025年第2期68卷 387-405页

作者： Qiuling Tao JinXin Yu Xiangyu Mu Xue Jia Rongpei Shi Zhifu Yao Cuiping Wang Haijun Zhang Xingjun Liu School of Materials Science and Engineering and Institute of Materials Genome&Big DataHarbin Institute of TechnologyShenzhenShenzhen 518055China School of Computer Science&Technology Harbin Institute of TechnologyShenzhenShenzhen 518055China Advanced Institute for Materials Research(WPI-AIMR) Tohoku UniversitySendai 980-8577Japan Department of Materials Science and Engineering College of MaterialsXiamen UniversityXiamen 361005China State Key Laboratory of Advanced Welding and Joining Harbin Institute of TechnologyShenzhenShenzhen 518055China

Machine learning (ML) has been widely used todesign and develop new materials owing to its low computational cost and powerful predictive capabilities. In recentyears, the shortcomings of ML in materials science have gradually emerged, with a primary concern being the scarcity ofdata. It is challenging to build reliable and accurate ML modelsusing limited data. Moreover, the small sample size problemwill remain long-standing in materials science because of theslow accumulation of material data. Therefore, it is importantto review and categorize strategies for small-sample learningfor the development of ML in materials science. This reviewsystematically sorts the research progress of small-samplelearning strategies in materials science, including ensemblelearning, unsupervised learning, active learning, and transferlearning. The directions for future research are proposed, including few-shot learning, and virtual sample *** importantly, we emphasize the significance of embedding material domain knowledge into ML and elaborate on thebasic idea for implementing this strategy.

关键词： material design machine learning small sample size few-shot learning material domain knowledge

来源：评论

学校读者我要写书评

暂无评论

Optimizing Near-Field Communication for Industrial IoT: A Biologically-Inspired Approach to Security and Efficiency

引用

IEEE Transactions on Consumer Electronics 2025年

作者： Kumar, Updesh Jannu, Srikanth Agarwal, Krishna Kant Naveena, C. Kumar, Prakash Ali, Jabir Vidyarthi, Ankit Ajay Kumar Garg Engineering College Department of Computer Science and Engineering U.P Ghaziabad India Vaagdevi Engineering College Department of CSE Bollikunta Warangal506 005 India Galgotias University Department of CSE Greater Noida India SJB Institute of Technology Department of CSE Kengeri Bengaluru560 060 India Jaypee Institute of Information Technology Department of CSE and IT Noida India Bennett University Noida School of Computer Science Engineering and Technology India

As the Industrial Internet of Things (IIoT) continues to expand, the need for robust, efficient, and secure communication mechanisms has become critical. Near-Field Communication (NFC) offers a promising solution for IIoT devices that require low-power, short-range, and secure interactions. However, optimizing NFC for industrial environments presents challenges, particularly in terms of security vulnerabilities and communication efficiency. This paper proposes a biologically-inspired approach to enhancing NFC performance in IIoT systems. Drawing from natural processes such as immune responses, swarm intelligence, and neural network communication strategies, we design adaptive algorithms that dynamically adjust communication protocols and security measures based on contextual factors. The proposed solution leverages decentralized decision-making, self-healing mechanisms, and resource-efficient strategies to improve both the security and efficiency of NFC systems in industrial applications. Through simulation and real-world testing, we demonstrate significant improvements in energy consumption, communication reliability, and resistance to cyber threats. Our results highlight the potential of biologically-inspired approaches for optimizing NFC in IIoT environments, paving the way for more resilient, scalable, and intelligent industrial communication systems. © 1975-2011 IEEE.

关键词： Swarm intelligence

来源：评论

学校读者我要写书评

暂无评论

Multi-modal signal integration for enhanced sleep stage classification: Leveraging EOG and 2-channel EEG data with advanced feature extraction

引用

Artificial Intelligence in Medicine 2025年 166卷 103152-103152页

作者： Samaee, Mahdi Yazdi, Mehran Massicotte, Daniel Signal and Image Processing Laboratory School of Electrical and Computer Engineering Shiraz University Shiraz Iran Laboratory of Signal and System Integration Department of Electrical and Computer Engineering Université du Québec à Trois-Rivières Trois-Rivières Canada

This paper introduces an innovative approach to sleep stage classification, leveraging a multi-modal signal integration framework encompassing Electrooculography (EOG) and two-channel electroencephalography (EEG) data. We explore the utility of various feature extraction techniques, including Short-Time Fourier Transform (STFT), Wavelet Transform, and raw signal processing, alongside the utilization of neural networks as feature extractors. This unique combination allows us to harness the benefits of traditional feature extraction methods while capitalizing on the power of neural networks to enhance classification performance. Our comprehensive classifier evaluation encompasses a range of models, including Long Short-Term Memory (LSTM) networks and XGBoost. Remarkably, our results reveal exceptional performance with the XGBoost classifier, achieving an overall accuracy of 84.57 % and a macro-F1 score of 78.21 % on the Sleep-EDF expanded dataset, and an overall accuracy of 86.02 % and a macro-F1 score of 81.96 % on the ISRUC-Sleep dataset. Class-specific accuracies highlight its proficiency, particularly in detecting wake and N2 stages, solidifying its superiority among the classifiers tested. This amalgamation of feature sets, complemented by Principal Component Analysis (PCA) for dimensionality reduction, underscores its significance in yielding top-tier classification outcomes. The integration of traditional feature extraction methods with neural networks as feature extractors creates a robust and comprehensive system for sleep stage classification, offering the advantages of both approaches to enhance the accuracy and reliability of the results. © 2025

关键词： Discrete wavelet transforms

来源：评论

学校读者我要写书评

暂无评论

A Battery Swapping Service Management System Reliability-Based Design Concerning Station Information Delay

引用

IEEE Transactions on Consumer Electronics 2025年

作者： Li, Xinyu Cao, Yue Hu, Ziyi Zhang, Xu Liu, Zhi Lei, Xinhai Wuhan University School of Cyber Science and Engineering Wuhan430073 China Wuhan Institute of Technology Hubei Key Laboratory of Intelligent Robot Wuhan430250 China School of Computing Sciences Norwich United Kingdom University of Electro-Communications Department of Computer and Network Engineering Tokyo Japan Wuhan Public Resources Trading Center Wuhan China

Since the long duration associated with traditional plug-in charging modes inevitably causes range anxiety to electric vehicle (EV) drivers, the battery swapping technology has emerged as a promising alternative. It thus becomes imperative to develop a reliable battery swapping service system, efficiently managing the large-scale energy replenishing demands from EVs. Firstly, this paper considers drivers' trip destination to minimize total trip duration, and the expense paid for battery swapping service to save users' service cost. By optimizing the requirement of battery swapping service during peak-price period, the expenditure for energy supplement can be reduced. Secondly, concerning in practice that the transmission delay is a crucial factor impacting the overall network and system reliability, this paper further investigates the impact of station information delay between the global controller and battery swapping stations (BSSs). By characterizing the delay into two types, namely, low and large delays, the scenario based on a known delay type are investigated. Thirdly, delay cases with multiple BSSs are also taken into account and a mixed decision-making method is developed. Finally, simulation results indicate the desirable performance of proposed scheme in satisfying user demand and alleviating impact of station information delay. © 1975-2011 IEEE.

关键词： Decision making

来源：评论

学校读者我要写书评

暂无评论

Schur roots and tilting modules of acyclic quivers over commutative rings

arXiv

引用

arXiv 2025年

作者： Iyama, Osamu Kimura, Yuta Graduate School of Mathematical Sciences The University of Tokyo 3-8-1 Komaba Meguro-ku Tokyo153-8914 Japan Department of Mechanical Engineering and Informatics Faculty of Engineering Hiroshima Institute of Technology 2-1-1 Miyake Saeki-ku Hiroshima731-5143 Japan

Let Q be a finite acyclic quiver and AQ the cluster algebra of Q. It is well-known that for each field k, the additive equivalence classes of support tilting kQ-modules correspond bijectively with the clusters of AQ. The aim of this paper is to generalize this result to any ring indecomposable commutative Noetherian ring R, that is, the additive equivalence classes of 2-term silting complexes of RQ correspond bijectively with the clusters of AQ. As an application, for a Dynkin quiver Q, we prove that the torsion classes of mod RQ corresponds bijectively with the order preserving maps from Spec R to the set of clusters. Copyright © 2025, The Authors. All rights reserved.

关键词： Equivalence classes

来源：评论

学校读者我要写书评

暂无评论

A Multi Framework for Predicting Alzheimer’s disease using Machine Learning

A Multi Framework for Predicting Alzheimer’s disease using ...

引用

Data Science, Agents & Artificial Intelligence (ICDSAAI), International Conference on

作者： Virendrakumar Anna Dhotre Dhanraj Babu Sripelli Jagadish Kavyashree Nagarajaiah R Ramesh Tejashree Tejpal Moharekar Computer Engineering Nutan Maharashtra Institute of Engineering and Technology Pune India College of Nursing Bharati Vidyapeeth (Deemed to be University) Navi Mumbai School of CS & AI SR University Warangal Telangana Department of MCA Sri Siddhartha Institute of Technology Tumkur Karnataka Dept of Computer Science and Engineering Chennai Institute of Technology Chennai Tamil Nadu India Yashwantrao Chavan School of Rural Development Shivaji University Kolhapur Maharashtra India

ISBN: (数字)9798331537555

ISBN: (纸本)9798331537562

Deep-Brain-Stimulation (DBS) is a demonstrated rehabilitation for Alzheimer's Disease (AD), regularly ensuing in an upgrade of engine work. In any case, a few unfortunate incidental upshots can happen past DBS, which can deteriorate the personal satisfaction of the patient. In this way, the medical group needs to painstakingly choose patients over whom to carry out DBS. In past, there have been a few endeavors to associate pre-usable information and DBS experimental outcomes, with generally centered on the engine symptomatology. We design and implement a Machine Learning (ML) based strategy called multimodel pipeline alluded to as FlowModel, ready to anticipate an enormous number of DBS medical results for AD. It is made out of a Artificial Neural Network (ANN) to squeeze medical data, a picture handling strategy from the conventional to retrieve morphological biomarkers our of T1imaging, and a Support Vector Machine (SVM) to execute the regressions. We authenticated it on 256 AD patients who gone through a DBS. FlowModel illustrated association determinants as lofty as 0.82 and had the option to altogether foresee 63 out of 84 scores, outperforming an existing linear model. The quantity of metrics that are anticipated with this pre-employable data was likewise observed to be related with the quantity of patients with this data accessible, showing that the FlowModel technique is still effectively learning. We introduced a new, ML based pipeline to foresee an assortment of post-usable clinical results of DBS for AD patients. It considered different bio-markers, emerging from various information modalities, illustrating strong relationship coefficients for certain scores from pre-employable information as it were. It shows that numerous clinical results of DBS can be anticipated rationalist to the particular simulation boundaries, as FlowModel has been approved without such incitement associated data.

关键词： Support vector machines Measurement Reviews Pipelines Imaging Machine learning Artificial neural networks Biomarkers Alzheimer's disease Engines

来源：评论

学校读者我要写书评

暂无评论

Federated Learning for Diffusion Models

引用

IEEE Transactions on Cognitive Communications and Networking 2025年

作者： Peng, Zihao Wang, Xijun Chen, Shengbo Rao, Hong Shen, Cong The School of Mathematics and Computer Science Nanchang University Nanchang330000 China The School of Electronics and Information Technology Sun Yat-sen University Guangzhou510006 China The School of Software Nanchang University Nanchang330000 China The Charles L. Brown Department of Electrical and Computer Engineering University of Virginia CharlottesvilleVA22904 United States

Diffusion models are powerful generative models that can produce highly realistic samples for various tasks. Typically, these models are constructed using centralized, independently and identically distributed (IID) training data. However, in practical scenarios, data is often distributed across multiple clients and frequently manifests non-IID characteristics. Federated Learning (FL) can leverage this distributed data to train diffusion models, but the performance of existing FL methods is unsatisfactory in non-IID scenarios. To address this, we propose FEDDDPM—Federated Learning with Denoising Diffusion Probabilistic Models, which leverages the data generative capability of diffusion models to facilitate model training. In particular, the server uses well-trained local diffusion models uploaded by each client before FL training to generate auxiliary data that can approximately represent the global data distribution. Following each round of model aggregation, the server further optimizes the global model using the auxiliary dataset to alleviate the impact of heterogeneous data on model performance. We provide a rigorous convergence analysis of FEDDDPM and propose an enhanced algorithm, FEDDDPM+, to reduce training overheads. FEDDDPM+ detects instances of slow model learning and performs a one-shot correction using the auxiliary dataset. Experimental results validate that our proposed algorithms outperform the state-of-the-art FL algorithms on the MNIST, CIFAR10 and CIFAR100 datasets. © 2025 IEEE. All rights reserved.

关键词： Federated learning

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：