检索结果-内蒙古大学图书馆

35th IEEE International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC 2024

作者： Mitsiou, Nikos A. Bouzinis, Pavlos S. Diamantoulakis, Panagiotis D. Sarigiannidis, Panagiotis G. Karagiannidis, George K. MetaMind Innovations P.C Kozani50100 Greece University of Western Macedonia Department of Informatics and Telecommunications Engineering Kozani50100 Greece Aristotle University of Thessaloniki Department of Electrical and Computer Engineering Thessaloniki54124 Greece Artificial Intelligence & Cyber Systems Research Center Lebanon

ISBN: (纸本)9798350362244

Deep neural networks (DNNs) have demonstrated their efficacy in delivering accurate solutions to a range of optimization problems. However, in the context of wireless communications, the size of these problems may vary across adjacent time slots, due to fast changes in the networks' architecture, e.g., the number of users. It is essential to note that this time-varying dimensionality of optimization problems in wireless networks necessitates adjustments in the DNN architecture, resulting in different numbers of input and output nodes. To address this challenge, in our paper, optimization problems of varying size are treated as distinct tasks. To tackle these tasks, a multi-task learning (MTL) approach based on modular sharing is proposed. The multi-task approach consists of a DNN, which is used to extract the solutions for all the optimization problems, and a router which manages which nodes and layers of the input and output layer of the DNN to be used during the forward propagation of each task. Consequently, all tasks share common parameters of the DNN, while the DNN dynamically adjusts to the number of nodes of its output and input layers. Numerical results demonstrate the superiority of the suggested approach over zero-padding, which is the current solution for handling resource allocation problems of varying size. © 2024 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Channel Attention HalfU-Net for Skin Lesion Segmentation 1

Channel Attention HalfU-Net for Skin Lesion Segmentation

引用

1st International Conference on Advanced engineering and Technologies, ICONNIC 2023

作者： Prasetyo, Heri Mahdy, Abid Ammar Nadhif, Abrar Dwi Fairuz Guo, Jing-Ming Hsia, Chih-Hsien Department of Informatics Surakarta Indonesia National Taiwan Univ. Sci. Technolgy Department of Electrical Engineering Taipei Taiwan National Ilan University Dept. Computer Sci. Infor. Engineering Yilan Taiwan

ISBN: (纸本)9798350306484

U-shaped architecture have gain popularity in the image segmentation task, especially for biomedical image. Regardless of its good performance, U-shaped architecture generally has huge number of parameters. Considering the resource constraint in actual medical environment, the needs lightweight model is essential. In order to overcome this problem, we propose Channel Attention HalfU-Net. The proposed method implement the existing medical image segmentation model named HalfU-Net. HalfU-Net can reduce the computational cost by simplify the decoder part of the UNet. Channel attention is assigned between ghost model for every stage in encoder part to capture more relevant feature. This leads to performance improvement in skin lesion segmentation problem. Despite that our model achieve state-ofthe-arts performance by surpassing U-Net in aspects of DSC and IoU metrics by 0.86% and 3.48% respectively, the number of parameter of our proposed method is more than 200 times less then U-Net. © 2023 IEEE.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Addressing Domain Mismatch in Unsupervised Neural Machine Translation

IEEE Transactions on Audio, Speech and Language Processing

引用

IEEE Transactions on Audio, Speech and Language Processing 2025年 33卷 472-482页

作者： Youyuan Lin Rui Wang Chenhui Chu Graduate School of Informatics Kyoto University Kyoto Japan Department of Computer Science and Engineering Shanghai Jiao Tong University Shanghai China

Pretrained models have taken full advantage of monolingual corpora and achieved impressive results in training Unsupervised Neural Machine Translation (UNMT) models. However, when adapting UNMT models with in-domain monolingual corpora for domain-specific translation tasks, one of the languages may lack in-domain corpora, resulting in the unequal amount and proportion of in-domain monolingual corpora in each language. This problem situation is known as Domain Mismatch (DM). This study investigates the impact of DM in UNMT. We find that DM causes a translation quality disparity. That is, while in-domain monolingual corpora of a language can enhance the in-domain translation quality into that particular language, this enhancement cannot be generalized to the other language, and the translation quality into the other language remains deficient. To address this problem, we propose Domain-Aware Adaptation (DAA), which can be embedded in the vanilla UNMT model training process. By passing sentence-level domain information to the model during training and inference, DAA gives higher weight to in-domain data from open-domain corpora related to specific domains to alleviate domain mismatch. The experimental results on German-English and Romanian-English translation tasks specified in the IT, Koran, medical, and TED2020 domains demonstrate that DAA can efficiently exploit open-domain corpora to mitigate the quality disparity of translation caused by DM.

关键词： Training Adaptation models Translation Neural machine translation Data models Speech processing

来源：评论

学校读者我要写书评

暂无评论

Detection of Tuberculosis Based on X-Ray Image Using the Vision Transformer Method

Detection of Tuberculosis Based on X-Ray Image Using the Vis...

引用

International Conference on Information and Communications Technology (ICOIACT)

作者： Raihan Ahnaf Nafisaputra Tessy Badriyah Prasetyo Wibowo Fitri Setyorini Informatics and Computer Engineering Department Politeknik Elektronika Negeri Surabaya (PENS) Surabaya INDONESIA

ISBN: (数字)9798331536206

ISBN: (纸本)9798331536213

Tuberculosis, or TB, is an infectious disease caused by the Mycobacterium tuberculosis bacteria, which can attack the human body regardless of age and gender. This bacterium is a very strong bacillus, requiring a long time to treat. The types of radiological examinations commonly used today to detect tuberculosis are chest X-rays and CT scans. However, not all hospitals have CT scan machines due to their higher cost, while almost every hospital has an X-ray machine. This research aims to develop an assistive device with a high level of accuracy in detecting the disease through the analysis of X-ray images. The solution is to develop a tuberculosis detection assistive device based on X-ray images using the Vision Transformer classification model. The dataset used is obtained from four references and contains chest X-ray images categorized into four classes: Normal, Tuberculosis, Pneumonia, and Covid-19. This Vision Transformer model is designed to classify chest X-ray images to predict lung diseases according to the specified labels. From this research, the model can predict with an accuracy of up to 89%. The result of this research can help doctors in making decisions and also serve as a learning tool for detecting diseases in X-ray images.

关键词： Training computer vision Accuracy Tuberculosis Hospitals Computed tomography Predictive models Transformers X-ray imaging Biomedical imaging

来源：评论

学校读者我要写书评

暂无评论

Meta-learning in Healthcare: A Survey

引用

SN computer Science 2024年第6期5卷 1-34页

作者： Rafiei, Alireza Moore, Ronald Jahromi, Sina Hajati, Farshid Kamaleswaran, Rishikesan Department of Computer Science Emory University Atlanta GA United States Department of Mechatronics Engineering University of Tehran Tehran Iran School of Science and Technology University of New England Armidale NSW Australia Department of Biomedical Informatics Emory University School of Medicine Atlanta GA United States Department of Biomedical Engineering Georgia Institute of Technology Atlanta GA United States

As a subset of machine learning, meta-learning, or learning to learn, aims at improving the model’s capabilities by employing prior knowledge and experience. A meta-learning paradigm can appropriately tackle the conventional challenges of traditional learning approaches, such as insufficient number of samples, domain shifts, and generalization. These unique characteristics position meta-learning as a suitable choice for developing influential solutions in various healthcare contexts, where the available data is often insufficient, and the data collection methodologies are different. This survey discusses meta-learning broad applications in the healthcare domain to provide insight into how and where it can address critical healthcare challenges. We first describe the theoretical foundations and pivotal methods of meta-learning. We then divide the employed meta-learning approaches in the healthcare domain into two main categories of multi/single-task learning and many/few-shot learning and survey the studies. Finally, we highlight the current challenges in meta-learning research, discuss the potential solutions, and provide future perspectives on meta-learning in healthcare. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2024.

关键词： Automated medicine Clinical decision support system Few-shot learning Healthcare Meta-leaning Multi-task learning

来源：评论

学校读者我要写书评

暂无评论

CORRECTIVE MAINTENANCE TIME MODEL OF THE 2-PARALLEL CONFIGURATION OF THE EARTH STATION SYSTEM USING CURVE FITTING 51

CORRECTIVE MAINTENANCE TIME MODEL OF THE 2-PARALLEL CONFIGUR...

引用

51st International Conference on computers and Industrial engineering, CIE 2024

作者： Kamaruddin, Nur Fatin Syamimi Rahim, Nadirah Abdul Mansor, Mahayaudin M. Department of Electrical and Computer Engineering International Islamic University Malaysia Jalan Gombak Kuala Lumpur Malaysia School of Mathematical Sciences College of Computing Informatics and Mathematics Universiti Teknologi MARA Shah Alam Malaysia

An Earth station system serves as a key hub in the ever-changing environment of satellite communication technology, allowing uninterrupted data interchange between the Earth and orbiting satellites. These systems are essential for various applications like telecommunications, remote sensing, and global positioning. In this research, a 2-parallel configuration is chosen for the Earth station system design because the literature suggests that it has the lowest operational cost and is reliably acceptable if compared to other parallel configurations. However, the 2-parallel configuration comes with the issue of care and upkeep, especially in the context of corrective maintenance time, which is the aim of this study. Corrective maintenance, or repairing components and systems after failure, is critical to the Earth station system operation. This study uses curve fitting techniques by using the Monte Carlo method from MATLAB software to foresee system component failures and predict their behavior over time. The useful insights can be extracted from previous maintenance data by using curve fitting, allowing one to make informed decisions regarding when and how to do maintenance. The results demonstrate the curve fitting of the combination of the corrective maintenance for seven subsystems with the mean time between failures (MTBF) of 1, 3, 5, 7, and 10 years, and it is shown that the first-degree polynomial has the lowest Root Mean Square Error (RMSE), which is 1.35. Furthermore, the corrective maintenance time model is developed not only to analyse component failures but also to assist in deriving appropriate maintenance guidelines in the future that can help system engineers repair the failed subsystem. © 2024 computers and Industrial engineering. All rights reserved.

关键词： MATLAB

来源：评论

学校读者我要写书评

暂无评论

Anomaly Detection of Patient Data in Public Hospital used Internet of Things Sensors 7

Anomaly Detection of Patient Data in Public Hospital used In...

引用

7th International Conference on Inventive Computation Technologies, ICICT 2024

作者： Kadir, Evizal Abdul Putri, Lenni Dianna Rosa, Sri Listia Siswanto, Apri Assidiqi, Farhan Evizal, Muhammad Fikri Universitas Islam Riau Department of Informatics Engineering Pekanbaru28284 Indonesia Universitas Prima Indonesia Department of Public Health Medan20118 Indonesia Telkom University Department of Electrical Engineering Bandung40257 Indonesia Eng National Dong Hwa University Computer Science And Information Hualien County974 Taiwan

ISBN: (纸本)9798350359299

Increasing number of patients doing treatment in a public hospital become as issue for the management. Limited number of equipment to detect abnormal patient and support from medical staff. This research aims to detect on the anomaly detection of hospital patient data through the utilization of Internet of Things (IoT) sensors. With the increasing integration of IoT devices in healthcare settings, such as sensors monitoring vital signs like temperature, blood pressure, and heart rate, the need for robust anomaly detection mechanisms becomes imperative. The research explores advanced algorithms and machine learning techniques to identify irregular patterns or outliers in patient data, aiming to enhance the early detection of potential health issues or abnormalities. Leveraging the vast amount of data generated by IoT sensors in hospital environments, the research aims to contribute to the development of more efficient and accurate anomaly detection systems, ultimately improving the quality of patient care and facilitating timely intervention by healthcare professionals. The findings of this research have significant implications for the evolving landscape of healthcare technologies, emphasizing the importance of ensuring data integrity and patient safety in the era of IoT-driven healthcare solutions. Patient data shows in graph to see how the anomaly indication for the patient. Multi sensor with parallel system will develop in future research plan. © 2024 IEEE.

关键词： Hospitals

来源：评论

学校读者我要写书评

暂无评论

A Machine Learning Approach to User Assignment in 5G Networks

A Machine Learning Approach to User Assignment in 5G Network...

引用

2022 Global Information Infrastructure and Networking Symposium, GIIS 2022

作者： Kouris, Michael Kokkinos, Vasileios Gkamas, Apostolos Pouyioutas, Philippos Bouras, Christos University of Patras Computer Engineering and Informatics Department Patras Greece University Ecclesiastical Academy of Vella Ioannina Greece University of Nicosia Computer Science Department Nicosia Cyprus

ISBN: (数字)9781665490955

ISBN: (纸本)9781665490955

With the advancement of wireless networks the data needs of the wireless internet have become so great, and the use of 4G and 5G so ubiquitous, that challenges arise in the availability and distribution of resources. In this paper we examine an application of Machine Learning and subsequently Neural Networks to offer a solution to this problem. The, now more than ever, profound availability of processing power greatly empowers and makes the deployment of these tools easier than ever, even in problems where in the past their application would not be feasible. The use of machine learning techniques to distribute resources in wireless networks is investigated and contrasted to a traditional algorithmic method. Scenarios are also entertained in which such approaches might be applicable. © 2022 IEEE.

关键词： Wireless networks

来源：评论

学校读者我要写书评

暂无评论

Runtime Verification of Autonomous Systems Utilizing Digital Twins as a Service 5

Runtime Verification of Autonomous Systems Utilizing Digital...

引用

5th IEEE International Conference on Autonomic Computing and Self-Organizing Systems Companion, ACSOS-C 2024

作者： Kristensen, Morten Haahr Bonizzi, Alberto Gomes, Claudio Hansen, Simon Thrane Isasa, Carlos Iven, Hannes Kamburjan, Eduard Larsen, Peter Gorm Leucker, Martin Talasila, Prasad Tang, Valdemar Trojgard Tonetta, Stefano Vosteen, Lars B. Wright, Thomas Aarhus University Department of Electrical and Computer Engineering Denmark Fondazione Bruno Kessler Italy University of Luxembourg Interdisciplinary Centre for Security Reliability and Trust Luxembourg Institute for Software Engineering and Programming Languages Universität zu Lübeck Germany University of Oslo Department of Informatics Norway

ISBN: (纸本)9798350389760

Autonomous Systems (AS) enable systems to adapt to drastic and unprecedented environmental changes, a capability that can be enhanced through the utilization of Digital Twins (DTs). However, the additional capabilities of AS come at the cost of explainability, as the expanding adaptation space complicates the reasoning about the system's behavior. For certain types of systems, it is crucial to ensure that specific properties are upheld despite the system's autonomous behavior. To facilitate the monitoring of these properties, we propose the use of Runtime Verification (RV). This tutorial demonstrates the integration of RV tools into the Digital Twins as a Service (DTaaS) platform to monitor and verify the behavior of AS in real-time. By exploring various methods to incorporate RV tools within a DT context, the tutorial aims to advance the application of RV technologies in autonomic computing and self-adaptive system design. Specifically, we demonstrate how the behavior of a self-configuring DT can be verified utilizing RV. This is accomplished through the DTaaS platform, which supports seamless deployment of DT-based AS. © 2024 IEEE.

关键词： Adaptive systems

来源：评论

学校读者我要写书评

暂无评论

Co-synthesis of Histopathology Nuclei Image-Label Pairs using a Context-Conditioned Joint Diffusion Model

arXiv

引用

arXiv 2024年

作者： Min, Seonghui Oh, Hyun-Jic Jeong, Won-Ki College of Informatics Department of Computer Science and Engineering Korea University Seoul Korea Republic of

In multi-class histopathology nuclei analysis tasks, the lack of training data becomes a main bottleneck for the performance of learning-based methods. To tackle this challenge, previous methods have utilized generative models to increase data by generating synthetic samples. However, existing methods often overlook the importance of considering the context of biological tissues (e.g., shape, spatial layout, and tissue type) in the synthetic data. Moreover, while generative models have shown superior performance in synthesizing realistic histopathology images, none of the existing methods are capable of producing image-label pairs at the same time. In this paper, we introduce a novel framework for co-synthesizing histopathology nuclei images and paired semantic labels using a context-conditioned joint diffusion model. We propose conditioning of a diffusion model using nucleus centroid layouts with structure-related text prompts to incorporate spatial and structural context information into the generation targets. Moreover, we enhance the granularity of our synthesized semantic labels by generating instance-wise nuclei labels using distance maps synthesized concurrently in conjunction with the images and semantic labels. We demonstrate the effectiveness of our framework in generating high-quality samples on multi-institutional, multi-organ, and multi-modality datasets. Our synthetic data consistently outperforms existing augmentation methods in the downstream tasks of nuclei segmentation and classification. © 2024, CC BY-NC-SA.

关键词： Tissue

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：