检索结果-内蒙古大学图书馆

A Model for Predicting Chronic Kidney Diseases Based on Medical Data Using Reinforcement Learning

SN computer science 2024年第4期5卷 353页

作者： Nramban Kannan, Senthil Kumar Aseervatham, Joshi Moholkar, Kavita Palanimuthu, Mithun Marappan, Saranya Narendran, Muthusamy Sathar, Banu Sengan, Sudhakar Department of Computer Science & amp Department of Artificial Intelligence and Data Science Panimalar Engineering College Tamil Nadu Chennai 600123 India Department of Computer Science and Business Systems Engineering JSPM’S Rajarshi Shahu College of Engineering Savitribai Phule Pune University Pune 411033 India Department of Computer Science and Engineering St. Joseph’s Institute of Technology Tamil Nadu Chennai 600119 India Department of Information Technology Paavai Engineering College Tamil Nadu Namakkal 637018 India Department Information Technology KCG College of Technology Tamil Nadu Chennai 600097 India Department of Master of Computer Application T. J. Institute of Technology Chennai 600100 India Department of Computer Science and Engineering PSN College of Engineering and Technology Tamil Nadu Tirunelveli 627451 India

Kidney diseases (KD) are a global public health concern affecting millions. Early detection and prediction are crucial for effective treatment. Artificial intelligence (AI) techniques have been used in KDP to analyze past medical records, applying patients’ Electronic Medical Record (EHR) data. However, conventional statistical analysis methods conflict with fully comprehending the complexity of EHR data. AI algorithms have helped early KDP learn and identify complex data patterns. However, challenges include training heterogeneous historical data, protecting privacy and security, and developing monitoring system regulations. This study addresses the primary challenge of training heterogeneous datasets for real-world evaluation. Early detection and diagnosis of chronic kidney disease (CKD) is crucial for improved outcomes, reduced healthcare costs, and reliable treatment. Early treatments are crucial for CKD, as it often develops without apparent symptoms. Predictive models, particularly those using reinforcement learning (RL), can identify significant trends in complex healthcare information, which standard techniques may struggle with. The study makes KDP more accurate and reliable using RL methods on clinical data. This lets doctors find diseases earlier and treat them better by looking at static and changing health measurements. Machine learning (ML) algorithms can enhance the accuracy of AI systems over time, enhancing their effectiveness in detecting and diagnosing diseases. In the current investigation, the RL-ANN model is implemented for performing enforceable CKD by assessing the outcomes of multiple neural networks, which include FNN, RNN, and CNN, according to parameters such as accuracy, sensitivity, specificity, prediction error, prediction rate, and kidney failure rate (KFR). The recommended RL-ANN method has a lower failure rate of 70% based on the KFR data. Further, the proposed approach earned 95% in PR and 70% in analysis of errors. However, the RL

关键词： Alerting systems and diagnostic assistance Chronic kidney disease Machine learning Medical history RL-ANN

来源：评论

学校读者我要写书评

暂无评论

FT-AED: Benchmark Dataset for Early Freeway Traffic Anomalous Event Detection 38

FT-AED: Benchmark Dataset for Early Freeway Traffic Anomalou...

引用

38th Conference on Neural Information Processing systems, NeurIPS 2024

作者： Coursey, Austin Ji, Junyi Quinones-Grueiro, Marcos Barbour, William Zhang, Yuhang Derr, Tyler Biswas, Gautam Work, Daniel B. Department of Civil and Environmental Engineering Vanderbilt University United States Institute for Software Integrated Systems United States Department of Computer Science United States

Early and accurate detection of anomalous events on the freeway, such as accidents, can improve emergency response and clearance. However, existing delays and mistakes from manual crash reporting records make it a difficult problem to solve. Current large-scale freeway traffic datasets are not designed for anomaly detection and ignore these challenges. In this paper, we introduce the first large-scale lane-level freeway traffic dataset for anomaly detection. Our dataset consists of a month of weekday radar detection sensor data collected in 4 lanes along an 18-mile stretch of Interstate 24 heading toward Nashville, TN, comprising over 3.7 million sensor measurements. We also collect official crash reports from the Tennessee department of Transportation Traffic Management Center and manually label all other potential anomalies in the dataset. To show the potential for our dataset to be used in future machine learning and traffic research, we benchmark numerous deep learning anomaly detection models on our dataset. We find that unsupervised graph neural network autoencoders are a promising solution for this problem and that ignoring spatial relationships leads to decreased performance. We demonstrate that our methods can reduce reporting delays by over 10 minutes on average while detecting 75% of crashes. Our dataset and all preprocessing code needed to get started are publicly released at https://***/ft-aed/to facilitate future research. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Challenges in Securing UAV IoT Framework: Future Research Perspective

引用

IEEE Transactions on Network and Service Management 2025年第3期22卷 2607-2629页

作者： Aljumah, Abdullah Ahanger, Tariq Ahamed Ullah, Imdad Prince Sattam Bin Abdulaziz University College of Computer Engineering and Sciences Al-Kharj11942 Saudi Arabia Prince Sattam Bin Abdulaziz University Department of Management Information Systems CoBA Al-Kharj11942 Saudi Arabia The University of Sydney School of Computer Science Faculty of Engineering SydneyNSW2006 Australia

Unmanned Aerial Vehicles (UAVs) offer the immense capability for allowing novel applications in a variety of domains including security, military, surveillance, medicine, and traffic monitoring. The prevalence of UAV systems is due to the collaboration and accomplishment of tasks efficiently and effectively. UAVs embedded with camcorders, GPS receivers, and wireless sensors propose enormous promise in realizing the Internet of Things (IoT) service delivery in vast domains. It results in establishing an airborne field of the IoT when empowered with communication protocols of LTE, 4G, and 5G/6G networks. However, numerous difficulties must be addressed before UAVs may be used effectively namely privacy, security, and administration. Conspicuously, in the current article, novel UAV-specific domains enabled by IoT and 5G/6G technology are explored. Moreover, the presented technique assesses sensor requirements and provides an overview of fleet management systems that address aerial networking, privacy, and security concerns. Furthermore, a framework based on the IoT-5G/6G aspect is proposed which can be deployed over UAVs. Finally, in a heterogeneous computational platform, the proposed framework provides a complete IoT architecture that enables secure UAVs. © 2004-2012 IEEE.

关键词： Unmanned aerial vehicles (UAV)

来源：评论

学校读者我要写书评

暂无评论

Introduction to the Minitrack on Learning Analytics 2023

Proceedings of the Annual Hawaii International Conference on...

引用

Proceedings of the Annual Hawaii International Conference on System sciences 2023年 2023-January卷 1269-1270页

作者： Willermark, Sara Islind, Anna Sigridur Deeva, Galina Óskarsdóttir, Mari University West School of Business Economics and IT Sweden Reykjavik University Department of Computer Science Iceland KU Leuven Research Centre for Information Systems Engineering Belgium

来源：评论

学校读者我要写书评

暂无评论

A coherent approach-based fine-tuning of segment anything model plus watershed algorithm for instance segmentation of mitochondria in electron microscopy images

引用

IEEE Access 2025年 13卷 98088-98105页

作者： Faska, Zahra Khrissi, Lahbib Mountasser, Imadeddine Haddouch, Khalid Akkad, Nabil El Alshathri, Samah El-Shafai, Walid Sidi Mohamed Ben Abdellah University Laboratory of Engineering Systems and Applications ENSA Fez Morocco Princess Nourah Bint Abdulrahman University College of Computer and Information Sciences Department of Information Technology P.O. Box 84428 Riyadh11671 Saudi Arabia Computer Science Department Riyadh11586 Saudi Arabia Menoufia University Faculty of Electronic Engineering Department of Electronics and Electrical Communications Engineering Menouf32952 Egypt

Increasingly available ultrastructural data from a continuously growing diversity of experimental conditions are driving new opportunities for fruitful neuroscientific hypotheses tested in intracellular compartments such as the nanoscale roles of, e.g., the mitochondria. Reliable morphological statistics are based on achieving highly accurate semantic segmentations of EM images. The state-of-the-art deep CNNs can be somewhat brittle;they tend to provide coarse and high-frequency-oscillatory solutions with discontinuities and false positives even for simple mitochondria segmentation. Historically, the current state-of-the-art in medical image segmentation would involve some variant of the encoder-decoder architecture, such as the U-Net architecture. The SAM does not perform as well, since it has not been explicitly trained for the task and does not demonstrate user-interactive, over one billion annotations mostly for natural images. However, the SAM may be applied to segment anything, including medical image segmentation challenging new datasets. This work is aimed at the difficult task of implementing domain adaptation in mitochondria segmentation within EM images obtained from various tissues and species, using deep learning. We do a systematic study to assess SAM's ability to perform segmentation in medical images, measure its performance on volumetric EM datasets, and show that it is powerful at segmenting instances even under challenging imaging conditions. We provide a fine-tuning SAM which can be naturally trained by SAM at an exemplary scale, benefiting from a diverse and large dataset over one million image masks in 11 modalities. This model would be able to perform precise segmentation for a wide range of targets under various imaging conditions, at the level of performance of specialized U-Net models, or even better. A visual comparison is shown between our fine-tuning SAM model and U-Net, along with an examination of different watershed post-processing st

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Deep learning and ensemble methods for anomaly detection in ICS security

引用

International Journal of Information Technology (Singapore) 2025年第3期17卷 1761-1775页

作者： Hossain, Md. Alamgir Hasan, Tahmid Karovic, jr, Vincent Abdeljaber, Hikmat A. M. Haque, Md. Alimul Ahmad, Sultan Zafar, Aasim Nazeer, Jabeen Mishra, B.K. Department of Computer Science and Engineering State University of Bangladesh South Purbachal Kanchan Dhaka 1461 Bangladesh Department of Computer Science and Engineering Prime University Dhaka Bangladesh Department of Information Systems Faculty of Management Comenius Bratislava Odbojárov 10 Bratislava 25 82005 Slovakia Department of Computer Science Faculty of Information Technology Applied Science Private University Amman 11931 Jordan MEU Research Unit Middle East University Amman 11831 Lebanon Department of Computer Science Veer Kunwar Singh University Bihar Ara India Department of Computer Science College of Computer Engineering and Sciences Prince Sattam Bin Abdulaziz University Alkharj 11942 Saudi Arabia University Center for Research and Development (UCRD) Department of Computer Science and Engineering Chandigarh University Gharuan Punjab Mohali 140413 India Department of Computer Science Aligarh Muslim University U.P Aligarh 202002 India P.G. Department of Physics Veer Kunwar Singh University Bihar Ara India

This research addresses the escalating threats to industrial control systems by introducing a novel approach that combines deep learning for feature selection with a robust ensemble-based classification technique to enhance anomaly detection. Our method utilizes a tailored autoencoder architecture to efficiently select features, followed by a Random Forest classifier to ensure reliable and generalizable detection of anomalies. Evaluated on the HAI Security Dataset, the proposed approach demonstrates exceptional performance, with significant improvements in detecting anomalous activities in ICS. The results highlight the method’s potential to enhance ICS security by providing a scalable and adaptable framework that evolves with emerging threats. This study not only advances cybersecurity methodologies for industrial systems but also lays the groundwork for future research, emphasizing the need for innovative and effective security solutions in the face of evolving cyber challenges. © Bharati Vidyapeeth's Institute of computer Applications and Management 2024.

关键词： Anomaly detection in ICS Cyber-resilience in critical infrastructure Deep learning for feature selection Industrial control systems security Random forest classifier for ICS

来源：评论

学校读者我要写书评

暂无评论

GDiffRetro: Retrosynthesis Prediction with Dual Graph Enhanced Molecular Representation and Diffusion Generation 39

GDiffRetro: Retrosynthesis Prediction with Dual Graph Enhanc...

引用

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Sun, Shengyin Yu, Wenhao Ren, Yuxiang Du, Weitao Liu, Liwei Zhang, Xuecang Hu, Ying Ma, Chen Department of Computer Science City University of Hong Kong Hong Kong Department of Computer Science and Engineering The Chinese University of Hong Kong Hong Kong Advance Computing and Storage Lab Huawei Technologies China Academy of Mathematics and Systems Science Chinese Academy of Sciences China Department of Electronic Engineering Tsinghua University China

ISBN: (纸本)157735897X

Retrosynthesis prediction focuses on identifying reactants capable of synthesizing a target product. Typically, the retrosynthesis prediction involves two phases: Reaction Center Identification and Reactant Generation. However, we argue that most existing methods suffer from two limitations in the two phases: (i) Existing models do not adequately capture the "face" information in molecular graphs for the reaction center identification. (ii) Current approaches for the reactant generation predominantly use sequence generation in a 2D space, which lacks versatility in generating reasonable distributions for completed reactive groups and overlooks molecules' inherent 3D properties. To overcome the above limitations, we propose GDiffRetro. For the reaction center identification, GDiffRetro uniquely integrates the original graph with its corresponding dual graph to represent molecular structures, which helps guide the model to focus more on the faces in the graph. For the reactant generation, GDiffRetro employs a conditional diffusion model in 3D to further transform the obtained synthon into a complete reactant. Our experimental findings reveal that GDiffRetro outperforms state-of-the-art semi-template models across various evaluative metrics. Copyright © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Convolutional Neural Network (CNN) for Fake Logo Detection: A Deep Learning Approach Using TensorFlow Keras API and Data Augmentation 3

Convolutional Neural Network (CNN) for Fake Logo Detection: ...

引用

3rd International Conference on Artificial Intelligence For Internet of Things, AIIoT 2024

作者： Murali, Saritha Gupta, Vaishnavi Saha, Soumyojyoti Sharan, Harsh Sinha, Prajwal Baibhav, Kumar Vellore Institute of Technology Department of Database Systems Vellore India Vellore Institute of Technology School of Computer Science and Engineering Vellore India

ISBN: (纸本)9798350372120

In numerous applications, including image retrieval, brand monitoring, and counterfeit identification, the detection and categorization of logos play pivotal roles. This study proposes a novel methodology employing convolutional neural networks (CNNs) for discerning fraudulent logos. The CNN's structure involves several convolutional layers at its core, accompanied by max-pooling layers to extract features for classification. Training the CNN model involves a varied dataset containing both authentic and fake logo images, and its effectiveness is assessed using well-established metrics like accuracy, precision, recall, and F1-score. The results of the study indicate that the proposed model is proficient in distinguishing counterfeit logos accurately. Furthermore, the study delves into potential applications and future directions of using CNN for identifying counterfeit logos. This research presents a robust methodology for counterfeit logo detection, thereby supporting endeavors aimed at curtailing the proliferation of counterfeit goods and upholding brand integrity. © 2024 IEEE.

关键词： Image classification

来源：评论

学校读者我要写书评

暂无评论

Categorizing methods for integrating machine learning with executable specifications

引用

science China(Information sciences) 2024年第1期67卷 5-19页

作者： David HAREL Raz YERUSHALMI Assaf MARRON Achiya ELYASAF Department of Computer Science and Applied Mathematics Weizmann Institute of Science Department of Software and Information Systems Engineering Ben-Gurion University of the Negev

Deep learning(DL),which includes deep reinforcement learning(DRL),holds great promise for carrying out real-world tasks that human minds seem to cope with quite *** promise is already delivering extremely impressive results in a variety of ***,while DL-enabled systems achieve excellent performance,they are far from *** has been demonstrated,in several domains,that DL systems can err when they encounter cases they had not hitherto ***,the opacity of the produced agents makes it difficult to explain their behavior and ensure that they adhere to various requirements posed by human *** the other end of the software development spectrum of methods,behavioral programming(BP) facilitates orderly system development using self-standing executable modules aligned with how humans intuitively describe desired system *** this paper,we elaborate on different approaches for combining DRL with BP and,more generally,machine learning(ML) with executable specifications(ES).We begin by defining a framework for studying the various approaches,which can also be used to study new emerging approaches not covered *** then briefly review state-of-the-art approaches to integrating ML with ES,continue with a focus on DRL,and then present the merits of integrating ML with *** conclude with guidelines on how this categorization can be used in decision making in system development,and outline future research challenges.

关键词： machine learning artificial intelligence grey box learning domain knowledge rules behavioral programming deep reinforcement learning survey

来源：评论

学校读者我要写书评

暂无评论

Enhancing Retrieval Augmented Generation systems Using AI Models and Graph Databases 7

Enhancing Retrieval Augmented Generation Systems Using AI Mo...

引用

7th IEEE International Conference on Emerging Smart Computing and Informatics, ESCI 2025

作者： Mukherjee, Souvik Shabnam, Soosan Hasan, Sara Ajitha, D. Department of Computer Science and Engineering Vellore Institute of Technology Vellore India Department of Software Systems School of Computer Science and Engineering (SCOPE) Vellore Institute of Technology Tamil Nadu Vellore India

ISBN: (纸本)9798331515683

This paper outlines the process of generating a Neo4j graph database powered by Language Models (LLMs). The primary goal is to extract structured information from unstructured data, including user profiles, paper briefs, and Slack messages, and convert them into Cypher queries. The data is then ingested into Neo4j to build a graph database that captures relationships between users, paper, technologies, and messages. A pipeline was developed to automate the process, ensuring accurate entity and relationship extraction using predefined templates. This approach allows for efficient data representation and supports consultancy in managing large datasets by generating insightful visualizations and querying capabilities. © 2025 IEEE.

关键词： Cypher queries data pipeline entity extraction graph database LLM Neo4j paper briefs Slack messages

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：