检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Islam, Sheikh Mohammed Shariful Abrar, Moloud Tegegne, Teketo Loranjo, Liliana Karmakar, Chandan Awal, Md Abdul Hossain, Md. Shahadat Kabir, Muhammad Ashad Mahmud, Mufti Khosravi, Abbas Siopis, George Moses, Jeban C. Maddison, Ralph Institute for Physical Activity and Nutrition Deakin University Melbourne Australia School of Information Technology Deakin University Melbourne Australia Westmead Applied Research Centre The University of Sydney Sydney Australia School of Electrical Engineering and Computer Science The University of Queensland Brisbane Australia School of Computer Data and Mathematical Sciences Western Sydney University Sydney Australia School of Computing Mathematics and Engineering Charles Sturt University Bathurst Australia Department of Computer Science Nottingham Trent University Nottingham United Kingdom Faculty of Medicine Health and Human Sciences Macquarie University Sydney Australia

Background: Machine learning models have the potential to identify cardiovascular diseases (CVDs) early and accurately in primary healthcare settings, which is crucial for delivering timely treatment and management. Although population-based CVD risk models have been used traditionally, these models often do not consider variations in lifestyles, socioeconomic conditions, or genetic predispositions. Therefore, we aimed to develop machine learning models for CVD detection using primary healthcare data, compare the performance of different models, and identify the best models. Methods: We used data from the UK Biobank study, which included over 500,000 middle-aged participants from different primary healthcare centers in the UK. Data collected at baseline (2006-2010) and during imaging visits after 2014 were used in this study. Baseline characteristics, including sex, age, and the Townsend Deprivation Index, were included. Participants were classified as having CVD if they reported at least one of the following conditions: heart attack, angina, stroke, or high blood pressure. Cardiac imaging data such as electrocardiogram and echocardiography data, including left ventricular size and function, cardiac output, and stroke volume, were also used. We used 9 machine learning models (LSVM, RBFSVM, GP, DT, RF, NN, AdaBoost, NB, and QDA), which are explainable and easily interpretable. We reported the accuracy, precision, recall, and F-1 scores;confusion matrices;and area under the curve (AUC) curves. Results: RBFSVM, GP, DT, and AdaBoost were the best performing models (accuracy: 96), and LSVM showed the weakest prediction performance (accuracy: 0.69). While RBFSVM and GP could significantly classify people with CVD better, LSVM and AdaBoost could classify healthy participants more accurately. The LSVM has a better false positive rate but a worse false negative rate. Moreover, the RBFSVM and GP have better performance in terms of false negatives. RBFSVM, GP, DT, NN, and AdaB

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Classification of Subtypes of Luminal Breast Cancer from Immunohistochemistry Slides 5th

Deep Learning Classification of Subtypes of Luminal Breast C...

引用

5th International Conference on computing and Network Communications, CoCoNet 2023

作者： Shaji, Vishnu Sharma, Mridul Mathew, Merin Malavika, K. Louis, Dhanya Mary Vijaykumar, Dehannathparambil Kottarathil Nair, Lakshmi Malavika George, Archana Gutjahr, Georg Department of Computer Science and Engineering Amrita School of Computing Amrita Vishwa Vidyapeetham Amritapuri India Department of Mathematics Amrita Vishwa Vidyapeetham Amritapuri Kerala Kollam India Department of Pharmacy Practice Amrita School of Pharmacy Amrita Institute of Medical Sciences And Research Centre Kerala Kochi India Department of Breast Oncology Amrita Institute of Medical Sciences And Research Center Amrita Vishwa Vidyapeetham Kerala Kochi India Department of Pathology Amrita Institute of Medical Sciences And Research Center Amrita Vishwa Vidyapeetham Kerala Kochi India Amrita Vishwa Vidyapeetham Amritapuri Kerala Kollam India

ISBN: (纸本)9789819747108

This paper proposes deep-learning models to assist in classifying subtypes of luminal breast cancer from immunohistochemistry slides. The classification has traditionally been done by pathologists who evaluate Ki67 indexes by manually counting positive and negative cells that are dyed in different colors in the immunohistochemistry slides. This manual process is error-prone and time-consuming. Therefore, this paper explores the possibility of automatically classifying immunohistochemistry slides. The performances of 4 pre-trained, deep convolutional networks are compared. Based on the good performance of the models, we conclude that the task of Ki67 classification for determining subtypes of luminal breast cancer is well-suited for deep learning models and that for this reason, manual counting of Ki67-positive cells are neither needed nor desirable. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

An authentication and plausibility model for big data analytic under LOS and NLOS conditions in 5G-VANET

引用

science China(Information sciences) 2020年第12期63卷 67-83页

作者： S.A.SOLEYMANI M.H.ANISI A.Hanan ABDULLAH M.Asri NGADI Sh.GOUDARZI M.Khurram KHAN M.Nazri KAMA School of Computing Faculty of Engineering Universiti Teknologi Malaysia (UTM) School of Computer Science and Electronic Engineering University of Essex Centre of Artificial Intelligence National University of Malaysia (UKM) Center of Excellence in Information Assurance (CoEIA) King Saud University Advanced Informatics School Menara Razak Universiti Teknologi Malaysia (UTM)

The exchange of correct and reliable data among legitimate nodes is one of the most important challenges in vehicular ad hoc networks(VANETs). Malicious nodes and obstacles, by generating inaccurate information, have a negative impact on the security of 5G-VANET. The big data generated in the vehicular network is also an issue in the security of VANET. To this end, a security model based on authentication and plausibility is proposed to improve the safety of network named ‘AFPM’. In the first layer, an authentication mechanism using edge nodes along with 5G is proposed to deal with the illegitimate nodes who enter the network and broadcast wrong information. In the authentication mechanism, because of the growth of the connected vehicles to the edge nodes that lead to generating big data and hence the inappropriateness of the traditional data structures, cuckoo filter, as a space-efficient probabilistic data structure, is used. In the second layer, a plausibility model by performing fuzzy logic is presented to cope with inaccurate information. The plausibility model is based on detection of inconsistent data involved in the event message. The plausibility model not only tackles with inaccurate, incomplete, and inaccuracy data but also deals with misbehaviour nodes under both line-of-sight(LOS) and non-line-of-sight(NLOS) conditions. All obtained results are validated through well-known evaluation measures such as F-measure and communication overhead. The results presented in this paper demonstrate that the proposed security model possesses a better performance in comparison with the existing studies.

关键词： authentication plausibility fuzzy logic cuckoo filter 5G-VANET big data

来源：评论

学校读者我要写书评

暂无评论

Shallow quantum circuits for efficient preparation of Slater determinants and correlated states on a quantum computer

引用

Physical Review A 2023年第2期108卷 022416-022416页

作者： Chong Hian Chee Daniel Leykam Adrian M. Mak Dimitris G. Angelakis Centre for Quantum Technologies National University of Singapore 3 Science Drive 2 Singapore 117543 Institute of High Performance Computing Agency for Science Technology & Research (A*STAR) 1 Fusionopolis Way 16-16 Connexis Singapore 138632 School of Electrical and Computer Engineering Technical University of Crete 73100 Chania Greece AngelQ Quantum Computing 531A Upper Cross Street 04-95 Hong Lim Complex Singapore 051531

Fermionic Ansatz state preparation is a critical subroutine in many quantum algorithms such as the variational quantum eigensolver for quantum chemistry and condensed-matter applications. The shallowest circuit depth needed to prepare Slater determinants and correlated states to date scales at least linearly with respect to the system size N. Inspired by data-loading circuits developed for quantum machine learning, we propose an alternate paradigm that provides shallower, yet scalable, O(dlog22N) two-qubit gate-depth circuits to prepare such states with d fermions, offering a subexponential reduction in N over existing approaches in second quantization, enabling high-accuracy studies of d≪O(N/log22N) fermionic systems with larger basis sets on near-term quantum devices.

关键词： Quantum algorithms for chemical calculations

来源：评论

学校读者我要写书评

暂无评论

Balancing Privacy and Performance: A Many-in-One Approach for Image Anonymization 39

Balancing Privacy and Performance: A Many-in-One Approach fo...

引用

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Jia, Xuemei Du, Jiawei Wei, Hui Xue, Ruinian Wang, Zheng Zhu, Hongyuan Chen, Jun National Engineering Research Center for Multimedia Software School of Computer Science Wuhan University China Hubei Key Laboratory of Multimedia and Network Communication Engineering Wuhan University China Centre for Frontier AI Research (CFAR) Institute of High Performance Computing (IHPC) A*STAR Singapore Centre for Frontier AI Research (CFAR) Institute for Infocomm Research (I2R) A*STAR Singapore

ISBN: (纸本)157735897X

The effective utilization of data through Deep Neural Networks (DNNs) has profoundly influenced various aspects of society. The growing demand for high-quality, particularly personalized, data has spurred research efforts to prevent data leakage and protect privacy in recent years. Early privacy-preserving methods primarily relied on instance-wise modifications, such as erasing or obfuscating essential features for de-identification. However, this approach highlights an inherent trade-off: minimal modification offers insufficient privacy protection, while excessive modification significantly degrades task performance. In this paper, we propose a novel Recombining for Obfuscation (FRO) approach to address this trade-off. Unlike existing methods that generate one anonymized instance by perturbing the original data on a one-to-one basis, our FRO approach generates an anonymized instance by reassembling mixed id-related features from multiple original data sources on a many-in-one basis. Instead of introducing additional noise for de-identification, our approach leverages the existing non-polluted features from other instances to anonymize data. Extensive experiments on identity identification tasks demonstrate that FRO outperforms previous state-of-the-art methods, not only in utility performance but also in visual anonymization. Copyright © 2025, Association for the Advancement of Artificial Intelligence.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Integrating Deep Learning with Fundus and Optical Coherence Tomography for Cardiovascular Disease Prediction

arXiv

引用

arXiv 2024年

作者： Maldonado-Garcia, Cynthia Zakeri, Arezoo Frangi, Alejandro F. Ravikumar, Nishant Centre for Computational Imaging and Simulation Technologies in Biomedicine School of Computing University of Leeds Leeds United Kingdom Division of Informatics Imaging and Data Sciences School of Health Sciences Faculty of Biology Medicine and Health University of Manchester Manchester United Kingdom School of Computer Science Faculty of Science and Engineering University of Manchester Kilburn Building Manchester United Kingdom Christabel Pankhurst Institute University of Manchester Manchester United Kingdom NIHR Manchester Biomedical Research Centre Manchester Academic Health Science Centre Manchester United Kingdom

Early identification of patients at risk of cardiovascular diseases (CVD) is crucial for effective preventive care, reducing healthcare burden, and improving patients’ quality of life. This study demonstrates the potential of retinal optical coherence tomography (OCT) imaging in conjunction with fundus photographs for identifying future adverse cardiac events. We utilized data from 977 patients who experienced CVD within a 5-year interval post-image acquisition, alongside 1,877 control participants without CVD, totaling 2,854 subjects. We propose a novel binary classification network based on a Multi-channel Variational Autoencoder (MCVAE), which learns a latent embedding of patients’ fundus and OCT images to classify individuals into two groups: those likely to develop CVD in the future and those who are not. Our model, trained on both imaging modalities, achieved promising results (AUROC 0.78 ± 0.02, accuracy 0.68 ± 0.002, precision 0.74 ± 0.02, sensitivity 0.73 ± 0.02, and specificity 0.68 ± 0.01), demonstrating its efficacy in identifying patients at risk of future CVD events based on their retinal images. This study highlights the potential of retinal OCT imaging and fundus photographs as cost-effective, non-invasive alternatives for predicting cardiovascular disease risk. The widespread availability of these imaging techniques in optometry practices and hospitals further enhances their potential for large-scale CVD risk screening. Our findings contribute to the development of standardized, accessible methods for early CVD risk identification, potentially improving preventive care strategies and patient outcomes. © 2024, CC BY-NC-ND.

关键词：

来源：评论

学校读者我要写书评

暂无评论

MIMO Virtual Array Design for mmWave 4D-Imaging Radar Sensors

MIMO Virtual Array Design for mmWave 4D-Imaging Radar Sensor...

引用

European Signal Processing Conference (EUSIPCO)

作者： Nazila Karimian Sichani Mohammad Alaee-Kerahroodi Ehsan Raei Bhavani Shankar M. R. Esfandiar Mehrshahi Seyyed Ali Ghorashi Department of Telecommunications Faculty of Electrical Engineering Shahid Beheshti University Tehran Iran Interdisciplinary Centre for Security Reliability and Trust (SnT) University of Luxembourg Luxembourg Department of Computer Science and Digital Technologies School of Architecture Computing and Engineering University of East London London UK

The recently developed 4D-Imaging mmWave MIMO radars have significant advantages over conventional radar sensors because they offer a large virtual array in both azimuth and elevation. However, the physical placement of transmit and receive antennas to achieve the desired virtual array, while considering cost and efficiency is not intuitive. In this paper, we propose an optimization framework for optimally placing the transmit and receive array physical elements. The proposed framework obtains a sequence of optimal placements based on the coordinate descent by defining a desired virtual array and number of transmit and receive antenna elements as the objective. The objective function is then monotonically minimized in each iteration of the proposed algorithm, converging to a solution that guarantees the desired number of transmit and receive antenna elements, while the obtained virtual array is as close to the desired virtual array as possible. The simulation results are validated by simulating the virtual array of multiple commercially available 4D-Imaging radar products.

关键词：

来源：评论

学校读者我要写书评

暂无评论

FCloudless: A Performance-Aware Collaborative Mechanism for JointCloud Serverless

FCloudless: A Performance-Aware Collaborative Mechanism for ...

引用

IEEE International Conference on Joint Cloud computing (JCC)

作者： Jianfei Liu Huaimin Wang Peichang Shi Yaojie Li Penghui Ma Guodong Yi National Key Laboratory of Parallel and Distributed Computing College of Computer Science National University of Defense Technology Changsha 410073 China Key Laboratory of Software Engineering for Complex Systems College of Computer Science National University of Defense Technology Changsha 410073 China Xiangjiang Lab Changsha 410073 China School of Advanced Interdisciplinary Studies Hunan University Of Technology and Business Changsha 410073 China

As a new stage in the development of the cloud computing paradigm, serverless computing has the high-level abstraction characteristic of shielding underlying details. This makes it extremely challenging for users to choose a suitable serverless platform. To address this, targeting the jointcloud computing scenario of heterogeneous serverless platforms across multiple clouds, this paper presents a jointcloud collaborative mechanism called FCloudless with cross-cloud detection of the full lifecycle performance of serverless platforms. Based on the benchmark metrics set that probe performance critical stages of the full lifecycle, this paper proposes a performance optimization algorithm based on detected performance data that takes into account all key stages that affect the performance during the lifecycle of a function and predicts the overall performance by combining the scores of local stages and dynamic weights. We evaluate FCloudless on AWS, AliYun, and Azure. The experimental results show that FCloudless can detect the underlying performance of serverless platforms hidden in the black box and its optimization algorithm can select the optimal scheduling strategy for various applications in a jointcloud environment. FCloudless reduces the runtime by 23.3% and 24.7% for cold and warm invocations respectively under cost constraints.

关键词：

来源：评论

学校读者我要写书评

暂无评论

InSAR for dielectric constant estimation of pavements: A feasibility study 14

InSAR for dielectric constant estimation of pavements: A fea...

引用

Earth Resources and Environmental Remote Sensing/GIS Applications XIV 2023

作者： Tessema, Tesfaye Gagliardi, Valerio Ciampoli, Luca Bianchini Tosti, Fabio School of Computing and Engineering University of West London St Mary s Road Ealing LondonW5 5RF United Kingdom The Faringdon Research Centre for Non-Destructive Testing and Remote Sensing University of West London St Mary s Road Ealing LondonW5 5RF United Kingdom Department of Civil Computer Science and Aeronautical Technologies Engineering Roma Tre University Via Vito Volterra 62 Rome Italy

ISBN: (纸本)9781510666979

Transport infrastructure health monitoring is fundamental to assure the road's integrity and the user's safety. Nowadays, pavement quality is conventionally assessed using different Non-Destructive Testing (NDT) methods. These methodologies are adequate but are limited in terms of spatial and temporal coverage. Depth estimation using InSAR data is not uncommon for application in arid areas, glaciers, and soil moisture studies. These areas are subject to substantial penetration at microwave frequencies. The methods assume the medium as a uniform volume with infinite depth and analyse the volume scattering of the radar wave. Determination of dielectric parameters of pavement layers and material inhomogeneities assessment is essential for pavement health monitoring. Here we examine the capability of InSAR data to estimate dielectric constant values in concrete pavements by interaction with the electromagnetic waves of radar satellite data. TerraSAR-X data are used to explore the dielectric constant estimation capabilities. The existing methods to invert for dielectric constant from the InSAR signal are applied to arid, ice-covered, and forested areas, but not pavements. This attempt will be the first of its kind to explore the large area coverage and with a frequent revisit of SAR satellites as a monitoring tool. © 2023 SPIE. All rights reserved.

关键词： Soil moisture

来源：评论

学校读者我要写书评

暂无评论

Aatma Yoga: Automation of Yoga Pose Recognition and Recommendation using Deep Learning

Aatma Yoga: Automation of Yoga Pose Recognition and Recommen...

引用

International Conference on Inventive Computation Technologies (ICICT)

作者： Vijaya Raghava Duppala Harika Yadav Marepalli Kirti Jain S Koduru Anusha Senthil Kumar Thangavel B Senthil Kumar Avadhani Bindu Latha Satish Jeeva Sekar Department of Computer Science and Engineering Amrita School of Computing Coimbatore Amrita Vishwa Vidyapeetham India Amrita Darshanam-International Centre For Spiritual Studies Coimbatore Amrita Vishwa Vidyapeetham India Research Department Krishnamacharya Yoga Mandiram Chennai India Prana Yoga Studio Chennai India

ISBN: (数字)9798350359299

ISBN: (纸本)9798350359305

Youngsters often face stress, pressure, and various challenges in today’s fast-paced world. The need for yoga arises as a powerful tool to alleviate stress, promote mental well-being, and enhance resilience. The purpose of this research is to present an innovative approach to personalized yoga practice by leveraging deep learning and computer vision techniques for real-time monitoring and correction of yoga poses. The system consists of mood-based yoga recommendation, yoga pose detection, yoga pose estimation, and pose correction. Through the seamless integration of Singular Value Decomposition (SVD), YOLOv3, PoseNet, and an angle heuristic algorithm, this work ensures accurate analysis and correction of the user’s yoga poses, facilitating an effective and personalized yoga experience. By addressing the lack of personalization and emotional oversight prevalent in traditional yoga practices, the work aims to promote emotional balance, stress relief, and physical wellness. Through this innovative approach, the work endeavors to transform the traditional paradigm of yoga practice, offering a more personalized, effective, and holistic wellness experience for practitioners.

关键词： YOLO Deep learning computer vision Mood Pose estimation Human factors Transforms

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：