检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Kalaie, Soodeh Bulpitt, Andy Frangi, Alejandro F. Gooya, Ali The School of Computing University of Leeds Leeds United Kingdom The Schools of Computer Science and Health Science The University of Manchester Manchester United Kingdom The School of Computing Science University of Glasgow Glasgow United Kingdom

Generative modelling for shapes is a prerequisite for In-Silico Clinical Trials (ISCTs), which aim to cost-effectively validate medical device interventions using synthetic anatomical shapes, often represented as 3D surface meshes. However, constructing AI models to generate shapes closely resembling the real mesh samples is challenging due to variable vertex counts, connectivities, and the lack of dense vertex-wise correspondences across the training data. Employing graph representations for meshes, we develop a novel unsupervised geometric deep-learning model to establish refinable shape correspondences in a latent space, construct a population-derived atlas and generate realistic synthetic shapes. We additionally extend our proposed base model to a joint shape generative-clustering multi-atlas framework to incorporate further variability and preserve more details in the generated shapes. Experimental results using liver and left-ventricular models demonstrate the approach’s applicability to computational medicine, highlighting its suitability for ISCTs through a comparative analysis. © 2024, CC BY-NC-ND.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

A Communication-efficient Conjunctive Query Scheme under Local Differential Privacy

A Communication-efficient Conjunctive Query Scheme under Loc...

引用

2024 IEEE Global Communications Conference, GLOBECOM 2024

作者： Zhang, Ellen Z. Guan, Yunguo Lu, Rongxing Zhang, Harry University of New Brunswick Faculty of Computer Science FrederictonE3B5A3 Canada Eastern Michigan University School of Information Security and Applied Computing YpsilantiMI48197 United States

ISBN: (纸本)9798350351255

Crowdsourcing has become a widely used method for data collection and analysis, yet its privacy remains a challenge. In this paper, we present a new efficient and privacy-preserving conjunctive query scheme for crowdsourcing scenarios. The scheme employs the Local Differential Privacy (LDP) technique to ensure both query privacy and high communication efficiency. Specifically, when an aggregator launches a conjunctive query to a set of crowdsourcing users, the query condition will not be leaked. To respond the query, each user just needs to return one bit back to the aggregator. By integrating prefix encoding technique, our proposed scheme can also efficiently support conjunctive queries with one range query condition. Detailed security analysis shows our proposed scheme can achieve desirable security requirements. In addition, performance evaluations also indicate its efficiency. Furthermore, extensive experiments demonstrate our proposed scheme can achieve high accuracy while ensuring ϵ-LDP. © 2024 IEEE.

关键词： Differential privacy

来源：评论

学校读者我要写书评

暂无评论

ResMHA-Net:Enhancing Glioma Segmentation and Survival Prediction Using a Novel Deep Learning Framework

引用

computers, Materials & Continua 2024年第10期81卷 885-909页

作者： Novsheena Rasool Javaid Iqbal Bhat Najib Ben Aoun Abdullah Alharthi Niyaz Ahmad Wani Vikram Chopra Muhammad Shahid Anwar Department of Computer Science School of Engineering and TechnologyIslamic University of Science and TechnologyAwantipora192122India Faculty of Computing and Technology Al-Baha UniversityAlaqiq65779-7738Saudi Arabia Research Groups in Intelligent Machines National School of Engineers of SfaxUniversity of SfaxSfax3038Tunisia Department of Electrical Engineering College of EngineeringKing Khalid UniversityAbha61421Saudi Arabia Department of Computer Science and Engineering Thapar Institute of Engineering and TechnologyPatiala147004Punjab Department of Electrical and Instrumentation Engineering Thapar Institute of Engineering and TechnologyPatiala147004Punjab Department of AI and Software Gachon UniversitySeongnam-si13120Republic of Korea

Gliomas are aggressive brain tumors known for their heterogeneity,unclear borders,and diverse locations on Magnetic Resonance Imaging(MRI)*** factors present significant challenges for MRI-based segmentation,a crucial step for effective treatment planning and monitoring of glioma *** study proposes a novel deep learning framework,ResNet Multi-Head Attention U-Net(ResMHA-Net),to address these challenges and enhance glioma segmentation ***-Net leverages the strengths of both residual blocks from the ResNet architecture and multi-head attention *** powerful combination empowers the network to prioritize informative regions within the 3D MRI data and capture long-range *** doing so,ResMHANet effectively segments intricate glioma sub-regions and reduces the impact of uncertain tumor *** rigorously trained and validated ResMHA-Net on the BraTS 2018,2019,2020 and 2021 ***,ResMHA-Net achieved superior segmentation accuracy on the BraTS 2021 dataset compared to the previous years,demonstrating its remarkable adaptability and robustness across diverse ***,we collected the predicted masks obtained from three datasets to enhance survival prediction,effectively augmenting the dataset *** features were then extracted from these predicted masks and,along with clinical data,were used to train a novel ensemble learning-based machine learning model for survival *** model employs a voting mechanism aggregating predictions from multiple models,leading to significant improvements over existing *** ensemble approach capitalizes on the strengths of various models,resulting in more accurate and reliable predictions for patient ***,we achieved an impressive accuracy of 73%for overall survival(OS)prediction.

关键词： Glioma MRI segmentation multihead attention survival prediction deep learning

来源：评论

学校读者我要写书评

暂无评论

Urban Slum Mapping Using Homogeneous Urban Patches 3

Urban Slum Mapping Using Homogeneous Urban Patches

引用

3rd IEEE International Conference on Intelligent Techniques in Control, Optimization and Signal Processing, INCOS 2024

作者： Prabhu, R. Priya, M. Sangeetha Dayana, R. Suresh, M. School of Computing Srm Institute of Science and Technology Department of Networking and Communications Kattankulathur Campus Chennai India Saveetha Engineering College Department of Computer Science and Engineering Thandalam Chennai India Jeppiaar Institute of Technology Department of Computer Science and Engineering Sriperumbudur Chennai India R.M.D. Engineering College Kavaraipettai Department of Mathematics Thiruvallur India

ISBN: (纸本)9798350361186

Slums (Informal Settlements) are the most dynamic areas whose geographical locations are inaccurate according to official statistics and maps because of the exponential expansion of urbanization strategies. As a result, a method has to be developed to distinguish between informal settlements and track their spatial patterns. This paper presents an integrated method for detecting slums from satellite images that employs homogeneous urban patches (HUP) and grey level co-occurrence matrix (GLCM)-based image segmentation. In order to extract meaningful spatial features, urban patches are segmented according to their homogeneity and GLCM analysis provides useful texture information about slum areas. The combination of both these GLCM and HUP features allows for the creation of a classification map that accurately identifies and maps slum areas. We utilized WorldView-2 (1.84m) satellite imagery of Madurai to test the effectiveness of proposed HUP approach. The validation process includes rigorous testing against ground truth data, with careful consideration given to the selection of training and testing datasets. The effectiveness of the proposed method is evaluated using metrics including overall accuracy, user accuracy, and producer accuracy Furthermore, the proposed algorithm outperformed existing GLCM-based approaches in terms of accuracy. © 2024 IEEE.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

A Prediction of Stock Time Series Data Using Single Layer Long Short-Term Memory 2

A Prediction of Stock Time Series Data Using Single Layer Lo...

引用

2nd IEEE International Conference on Data science and Network Security, ICDSNS 2024

作者： Al-Fatlawy, Ramy Riad Lande, Jayapal Sundaram, S. Meenakshi. Saranya, N. Naga Podaralla, Rakesh College of Technical Engineering The Islamic University Department of Computers Techniques Engineering Najaf Iraq School of Computer Science & Artificial Intelligence Sr University Warangal India Nitte Meenakshi Institute of Technology Department of Cse Bengaluru India Saveetha College of Liberal Arts & Science Simats Department of Computer Applications Chennai India School of Computing and Information Technology Reva University Bangalore India

ISBN: (纸本)9798350373110

Financial and cash flow data is a particular kind of past temporal sequence information which gives a huge quantity of data which is employed often in the task including data analysis. For investors and financial professionals, the major question is how to forecast stock time series data. Stock prices forecasting is a challenging task because of time series data volatility, non-linearity and significant noise on the value of stocks. The earlier study only focused on alone parameter stock called close cost, hence, Single Layer Long Short-term Memory (LSTM) model is proposed. This model suggests the contribution as stock information and it predicts two stocks characteristics namely high cost and close cost for the day after. The dataset for Shanghai composite index is utilized for conducting experiments and by the use of existing techniques the comparisons are done. From the results, it clearly shows that proposed single layer LSTM achieved MAE of 26.961, RMSE of 12.112 and R2 of 0.987 which is better when compared with the existing models. © 2024 IEEE.

关键词： Long short-term memory

来源：评论

学校读者我要写书评

暂无评论

Remote Sensing Data-Based Satellite Image Analysis in Water Quality Detection for Public Health Data Modelling

引用

Remote Sensing in Earth Systems sciences 2024年第4期7卷 532-541页

作者： Balakrishnan, S. Preetam Raj, P Michael Somasekar, J. Kumar, Kambala Vijaya Amutha, S. Sangeetha, A. Department of Computer Science and Engineering Aarupadai Veedu Institute of Technology (AVIT) Vinayaka Mission’s Research Foundation (Deemed to be University) Chennai India Department of Computer Science and Engineering MLR Institute of Technology Telangana Dundigal Hyderabad 500043 India Department of Computer Science and Engineering Faculty of Engineering and Technology JAIN (Deemed-to-be-University) Karnataka Bengaluru 562112 India Department of Computer Science and Engineering Koneru Lakshmaiah Education Foundation Vaddeswaram AP India Department of Computer Science and Engineering School of Computing Vel Tech Rangarajan Dr. Sagunthala R & amp D Institute of Science and Technology Tamilnadu Chennai India Department of Artificial Intelligence and Data Science Kongunadu College of Engineering and Technology TamilNadu Tholurpatti India

One of the most important aspects of environmental sustainability is water quality monitoring. The ecology is impacted by poor water quality in addition to aquatic life. With the volume of data being collected from satellite systems, the issue of effectively and automatically extracting water bodies is growing more pressing. One of the most important things to do for managing and safeguarding water resources is to conduct extensive monitoring of water quality parameters, or WQPs. It is still difficult to monitor optically inactive WQPs in inland waters, such as total nitrogen (TN), ammoniacal nitrogen (AN), and total phosphorus (TP). Water quality (WQ) classification and prediction can be greatly enhanced with the use of artificial intelligence (AI). This research proposes novel method in healthcare detection by water quality analysis based on classification utilising ML model. Input is collected as healthcare dataset based on water quality as well as processed for noise removal and normalisation. The processed data features are extracted and classified utilising reinforcement Gaussian stacked layer network with variational fuzzy Markov basis neural network. In order to suggest best course of action for the water bodies, it is imperative that any study look into the spatiotemporal changes of the dominant water quality parameters (WQPs). WQP concentrations have typically been determined by extensive fieldwork. These findings could help identify and regulate non-point source pollution, offer a large-scale spatial picture of the water quality, and pinpoint vulnerable locations and times of water pollution. Experimental analysis is carried out for various water sample dataset in terms of detection accuracy, mean precision, recall, F-1 score and AUC. The proposed technique obtained 97% of detection accuracy, 95% of recall, 91% of F-1 score, 96% of AUC, and 94% of MAP. © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2024.

关键词： Classification Fuzzy Markov basis Healthcare detection Machine learning model Water quality analysis

来源：评论

学校读者我要写书评

暂无评论

Smart Captures- An Advanced Attendance System using Facial Recognition 1

Smart Captures- An Advanced Attendance System using Facial R...

引用

1st International Conference on Cognitive computing and Engineering Education, ICCCEE 2023

作者： Gunjan, Reena Mital, Anannya Jain, Muskan Pangaonkar, Satyajit Shete, Virendra Rokade, Prakash MIT School of Computing MIT Art Design and Technology University Department of Computer Science Engineering Pune India School of Computer Science and Engineering Avantika University Madhya Pradesh Ujjain India MIT School of Engineering and Science MIT Art Design and Technology University Department of Electronics and Communication Engineering Pune India Cloud and Infrastructure Services Infosys Ltd Maharashtra Pune India

ISBN: (纸本)9798350332803

Facial Recognition is one of the most productive and important techniques in image processing. It uses Machine Learning and Artificial Intelligence techniques to identify, collect, store, and evaluate face characteristics so that they can be matched to photos of people in a database. Nowadays, many kinds of research are focusing on human detection mainly in the context of marking students' attendance. Marking attendance is a procedure of recognizing students' faces and marking their attendance for that class. The need for this application was generated because after the pandemic use of touch biometric systems was suspended in many places and the manual process of taking attendance and maintaining paper records was a very tedious job and this could be manipulated very easily by anyone. Some attendance systems with face recognition already got developed but they were very basic and only had the feature of just recognizing the face once and marking the attendance of that student. So, by bringing in some unique features, here, developed 'Smart Captures', an advanced attendance system for universities, schools, and education-based places. Smart Captures was entitled to capturing the students' movement from the classroom (in out) and marking their attendance according to the set constraints using IoT, thereafter, making it visible to the faculties on their mobile application. This provided the data analytics for regular attendance on the dashboard to the admin with the administrative power using the application. © 2023 IEEE.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

A Secure Communication Protocol for Unmanned Aerial Vehicles

引用

computers, Materials & Continua 2022年第1期70卷 601-618页

作者： Navid Ali Khan N.Z.Jhanjhi Sarfraz Nawaz Brohi Abdulwahab Ali Almazroi Abdulaleem Ali Almazroi School of Computer Science and Engineering(SCE) Taylor’s University Lake-side CampusSubang Jaya47500Malaysia School of Information and Technology Monash UniversityMalaysia University of Jeddah College of Computing and Information Technology at KhulaisDepartment of Information TechnologyJeddahSaudi Arabia Department of Computer Science Rafha Community CollegeNorthern Border UniversityArar91431Saudi Arabia

Mavlink is a lightweight and most widely used open-source communication protocol used for Unmanned Aerial *** UAVs and autopilot systems support it,and it provides bi-directional communication between the UAV and Ground Control *** communications contain critical information about the UAV status and basic control commands sent from GCS to UAV and UAV to *** order to increase the transfer speed and efficiency,the Mavlink does not encrypt the *** a result,the protocol is vulnerable to various security attacks such as Eavesdropping,GPS Spoofing,and *** this study,we tackle the problem and secure the Mavlink communication *** leveraging the Mavlink packet’s vulnerabilities,this research work introduces an experiment in which,first,the Mavlink packets are compromised in terms of security requirements based on our threat *** results show that the protocol is insecure and the attacks carried out are *** overcomeMavlink security,an additional security layer is added to encrypt and secure the *** encryption technique is proposed that makes the communication between the UAV and GCS *** results show that the Mavlink packets are encrypted using our technique without affecting the performance and *** results are validated in terms of transfer speed,performance,and efficiency compared to the literature solutions such as MAVSec and benchmarked with the original Mavlink *** achieved results have significant improvement over the literature and Mavlink in terms of security.

关键词： Unmanned aerial vehicles mavlink protocol drones security UAVs communication

来源：评论

学校读者我要写书评

暂无评论

Attention-based Multimodal Bilinear Feature Fusion for Lung Cancer Survival Analysis 23

Attention-based Multimodal Bilinear Feature Fusion for Lung ...

引用

23rd IEEE International Conference on Bioinformatics and Bioengineering, BIBE 2023

作者： Na, Hongbin Wang, Lilin Zhuang, Xinyao He, Jianfei Liu, Zhenyu Wang, Zimu Gan, Hong-Seng School of Computer Science and Engineering University of New South Wales Sydney Australia School of AI and Advanced Computing Xi'an Jiaotong-Liverpool University Suzhou China School of Advanced Technology Xi'an Jiaotong-Liverpool University Suzhou China

ISBN: (纸本)9798350393118

Survival analysis (SA) is an essential task that aims to predict survival status and duration, determine individual and precise treatment strategies, and assess disease intensity and direction. However, the current research on multimodal SA has identified three unique challenges: inefficient cross-modal information integration, insufficient inter-modal key features, and noisy data. In this paper, we propose a novel SA framework, named Attention-based Multimodal Bilinear Feature Fusion (AMBF)-SA, to address the aforementioned challenges. Specifically, AMBF-SA first performs feature extraction with the off-the-shelf models on each modality separately, then fuses the features between multiple sources and modalities using our proposed AMBF method, and finally outputs the survival prediction by a multi-layer perception (MLP). Experimental results on the Non-small Cell Lung Cancer (NSCLC) Radiogenomics dataset demonstrate remark performance of AMBF-SA compared with the rest of the experimented models, including the models trained with single and combined modalities under the Mean Absolute Error (MAE) and the Concordance Index (C-index) evaluation metrics, indicating the usefulness of our proposed framework. © 2023 IEEE.

关键词： Biological organs

来源：评论

学校读者我要写书评

暂无评论

Multi-focus image registration algorithm based on image preprocessing and improved SIFT

Multi-focus image registration algorithm based on image prep...

引用

2022 International Conference on computer Application and Information Security, ICCAIS 2022

作者： Lin, Ke Pan, Wei School of Computer Science China West Normal University Sichuan Nanchong637009 China Nanchong Key Laboratory of IoT Awareness and Distributed Computing Sichuan Nanchong637009 China

ISBN: (纸本)9781510663459

Image registration is an important technology in the image processing, which is widely used in computer vision, image fusion, image reconstruction and other fields. To solve the problem of poor image registration due to fuzzy regions in multi-focus images, this paper proposes a multi-focus image registration algorithm based on image preprocessing and improved SIFT. The reference image and the image to be registered are preprocessed by Gaussian smoothing. And then all images are feature extracted and classified. Last, the image is feature matched by bidirectional adaptive threshold algorithm, and finally the registered image is obtained by the registration result. Compared with SIFT algorithm, the number of feature extraction and feature matching of the proposed algorithm increase by 56.57% and 67.43% respectively. © 2023 SPIE.

关键词： Image fusion

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：