检索结果-内蒙古大学图书馆

Personalizing Federated Instrument Segmentation With Visual Trait Priors in Robotic Surgery

IEEE Transactions on Biomedical Engineering 2025年第6期72卷 1886-1896页

作者： Xu, Jialang Wang, Jiacheng Yu, Lequan Stoyanov, Danail Jin, Yueming Mazomenos, Evangelos B. University College London UCL Hawkes Institute The Department of Medical Physics and Biomedical Engineering London United Kingdom Xiamen University Department of Computer Science School of Informatics Xiamen China The University of Hong Kong Department of Statistics and Actuarial Science Hong Kong National University of Singapore Department of Biomedical Engineering The Department of Electrical and Computer Engineering Singapore

Personalized federated learning (PFL) for surgical instrument segmentation (SIS) is a promising approach. It enables multiple clinical sites to collaboratively train a series of models in privacy, with each model tailored to the individual distribution of each site. Existing PFL methods rarely consider the personalization of multi-headed self-attention, and do not account for appearance diversity and instrument shape similarity, both inherent in surgical scenes. We thus propose PFedSIS, a novel PFL method with visual trait priors for SIS, incorporating global-personalized disentanglement (GPD), appearance-regulation personalized enhancement (APE), and shape-similarity global enhancement (SGE), to boost SIS performance in each site. GPD represents the first attempt at head- wise assignment for multi-headed self-attention personalization. To preserve the unique appearance representation of each site and gradually leverage the inter-site difference, APE introduces appearance regulation and provides customized layer- wise aggregation solutions via hypernetworks for each site's personalized parameters. The mutual shape information of instruments is maintained and shared via SGE, which enhances the cross-style shape consistency on the image level and computes the shape-similarity contribution of each site on the prediction level for updating the global parameters. PFedSIS outperforms state-of-the-art methods with +1.51% Dice, +2.11% IoU, -2.79 ASSD, -15.55 HD95 performance gains. © 1964-2012 IEEE.

关键词： Robotic surgery

来源：评论

学校读者我要写书评

暂无评论

A Hybrid Recurrent Neural Network with Mini-Batch Gradient Descent Method for Multiple Sclerosis Disease Prediction

A Hybrid Recurrent Neural Network with Mini-Batch Gradient D...

引用

International Conference on Advances and Applications of Artificial Intelligence and Machine Learning, ICAAAIML 2023

作者： Kavipriya, E. Sasikala, S. Department of Computer Science With Cognitive Systems and AIML Hindusthan College of Arts & Science Tamil Nadu Coimbatore India

ISBN: (纸本)9789819795062

Machine Learning techniques on different sectors plays a vital role for prediction analysis. In healthcare sector, diagnosis support and prognostic evaluation has been done using various algorithms of machine learning. The Multiple Sclerosis (MS) is one of the chronic diseases which affects the central nervous system of human. There are multiple symptoms and disabilities indicate the issues of Multiple Sclerosis. This neurological disorder has various patterns and disease progression rate among worldwide. This causes a variety of symptoms including fatigue, numbness, visual difficulties, urinary incontinence, and difficulty moving and cognitive deficiencies. This paper analyzes the risk factors of Multiple Sclerosis and early prediction using Machine learning methods. This work proposed hybrid Recurrent Neural Network with mini-batch gradient descent method for Multiple Sclerosis prediction. The performance of this proposed method is evaluated and the achieved accuracy rate is 92%. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

RT-PPS: Real-time privacy-preserving scheme for cloud-hosted IoT data

引用

Journal of High Speed Networks 2025年第1期31卷 71-89页

作者： Elhoseny, Mohamed Riad, Khaled College of Computing and Informatics University of Sharjah Sharjah27272 United Arab Emirates Faculty of Computers and Information Mansoura University Mansoura35516 Egypt Computer Science Department College of Computer Sciences & Information Technology King Faisal University Al-Ahsa 31982 Saudi Arabia Mathematics Department Faculty of Science Zagazig University Zagazig44519 Egypt

The Internet of Things (IoT) is a rapidly growing network of devices that can communicate with each other and with cloud-based services. These devices generate vast amounts of data that can be used to provide valuable insights into user behavior, environmental conditions, and other important factors. However, as this data is collected and processed by cloud-hosted services, there is a growing concern about privacy and security. Without adequate protection, sensitive information could be exposed to hackers or other malicious actors, putting both individuals and organizations at risk. To address this challenge, real-time privacy-preserving techniques can be used to protect IoT data without compromising its value. This paper introduces an efficient Real-time privacy-preserving scheme (RT-PPS) for cloud-hosted IoT data. RT-PPS employs multi-authority attribute-based encryption on a hybrid cloud environment to keep data secure and private, while still allowing it to be processed and analyzed by cloud-hosted services. RT-PPS has efficient response time and resource consumption, which gives it the ability to handle a huge number of concurrent users at the same time without notable delay. The proposed RT-PPS has been validated through extensive experimental evaluation on a variety of configurations. Moreover, the proposed scheme has been computationally compared with the state-of-the-artwork. RT-PPS has shown excellent performance, effectiveness, and efficiency. The RT-PPS encryption time for a 1 GB dataset while considering 1024 slices is approximately 1000 ms. Also, the RT-PPS decryption time for a 1 GB ciphertext while considering 1024 slices are approximately 235 ms. Finally, RT-PPS is proven secure against any polynomial-time attacks and their variations that have at most a negligible advantage in the introduced security model. Moreover compared to most of the state-of-the-artwork, RT-PPS reduced the ciphertext size and lowered the computations in the encryption, key g

关键词： Ciphertext

来源：评论

学校读者我要写书评

暂无评论

A Novel Framework for Gauging Information Extracted from Smartphones Using Neutrosophic Logic

引用

Neutrosophic Sets and systems 2025年 76卷 154-171页

作者： Abou Alzahab, R.M. Ismail, Amr Abd Elkhalik, S.H. Shams, M.Y. El-Bakry, H.M. Salama, A.A. Department of Mathematics and Computer Science Faculty of Science Port Said University Egypt Department of Cybersecurity College of Engineering and Information Technology Buraydah Private Colleges Buraydah51418 Saudi Arabia Department. of Communication and Computer Engineering Air Defense University Egypt Faculty of Artificial Intelligence Kafrelsheikh University Kafrelsheikh33516 Egypt Department of Information Systems. Faculty of Computer and Information Science Mansoura University Egypt

Smartphones contain a vast amount of information about their users, which can be used as evidence in criminal cases. However, the sheer volume of data can make it challenging for forensic investigators to identify and use the most relevant information. Neutrosophic logic is a generalization of fuzzy logic that allows for uncertainty and vagueness, making it a more potent tool for dealing with the ambiguity of real-world data. The proposed framework aims to identify and utilize the most relevant information for forensic investigators, making it easier to solve criminal cases using Neutrosophic logic. This novel approach provides a more powerful tool for dealing with the ambiguity of smartphone data, ultimately improving the accuracy and effectiveness of forensic investigations. Our research has utilized Neutrosophic logic to evaluate the degree of truth, falsity, or Indeterminacy of this information. Additionally, this study analyzes conversations between individuals using Excel’s fuzzy lookup add-in to determine the percentage of truth and false in each conversation. The results were compiled into a dataset and utilized a Neutrosophic Python code to evaluate the information. The results indicate the percentage of truth, false, and Indeterminacy in each conversation and which can be used to determine its admissibility as evidence and which not. © 2025, University of New Mexico. All rights reserved.

关键词： Smartphones

来源：评论

学校读者我要写书评

暂无评论

A Mathematical and Dynamical Model for Analyzing H-Shaped PM Vernier Motor for Electric Motorcycle Mid-Drive Applications

引用

IEEE Transactions on Energy Conversion 2025年第1期40卷 529-543页

作者： Ghods, Mehrage Faiz, Jawad Bazrafshan, Mohammad Amin Gorginpour, Hamed Toulabi, Mohammad Sedigh University of Tehran School of Electrical and Computer Engineering College of Engineering Tehran*** Iran Persian Gulf University Department of Intelligent Systems Engineering and Data Science Bushehr*** Iran Oakland University Department of Electrical and Computer Engineering RochesterMI48309 United States

The permanent magnet (PM) Vernier machines enhance torque density and decrease cogging torque compared to conventional permanent magnet synchronous motor. This paper presents a novel fractional-slot H-shaped PM Vernier machines, which incorporates unique design for the PM housing and rotor core construction. These designs are specifically optimized for the small-scale electric vehicles application. The choice of the PM arrangement is based on the combination of V-shaped and spoke-array PM topologies. The aim is to achieve a higher torque density than V-shaped PM machines, while reducing cogging torque compared to spoke-array PM machines. By using this methodology, leakage fluxes in the terminal region of the PMs reduce, resulting in an augmentation of both flux-linkage and power factor. A prototype 680 W motor was fabricated with 12 slots and 8 poles to verify the accuracy and effectiveness of the proposed model. Mathematical model and dynamical analysis are proposed for the analytical prediction of machine performance. Due to the unique shape feature of the machine, the conformal mapping approach is applied to design the rotor core. To validate the model, the predicted performance of the motor was compared with those obtained by finite element analysis and experiments. © 1986-2012 IEEE.

关键词： Permanent magnets

来源：评论

学校读者我要写书评

暂无评论

Hybrid Deep Learning for Proactive Driver Risk Prediction and Safety Enhancement

Hybrid Deep Learning for Proactive Driver Risk Prediction an...

引用

2025 International Conference on Multi-Agent systems for Collaborative Intelligence, ICMSCI 2025

作者： Sakthivel, K. Kowsalya, A. Durgadevi, M. Dhaneswar, R.C. K.S.Rangasamy College of Technology Department of Computer Science and Business Systems Namakkal India

ISBN: (数字)9798331509828

ISBN: (纸本)9798331509828

A more advanced deep learning architecture that aims to improve in situ driver safety by perceiving the driving behavior and monitoring the surrounding road environment will be proposed in this study. It will employ data provided by in-car cameras and sensors to extract critical spatial features like nearby vehicles, pedestrians, and road signs in addition to temporal patterns reflecting vehicle movement, speed variations, and proximity to potential hazards. The framework integrates spatial and temporal data to identify unsafe driving behaviors such as lane deviations, abrupt speed changes, and unsafe following distances. It further predicts future risks and provides real-time alerts to the driver to prevent accidents. Extensive testing on real-world datasets demonstrates that the system significantly improves hazard detection accuracy, driver behavior monitoring, and performance with low latency, making it a highly practical solution for enhancing safety in modern vehicles. The system plays a vital role in promoting safer roadways and reducing the likelihood of accidents because it can process data in real time. © 2025 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Artificial Intelligence for Automated Support in Educational systems

Artificial Intelligence for Automated Support in Educational...

引用

2025 International Conference on Multi-Agent systems for Collaborative Intelligence, ICMSCI 2025

作者： Karthik, R. Maha Vishnu, S. Saran, R. Shashidharan, S. K.S.Rangasamy College of Technology Department of Computer Science and Business Systems Namakkal India

ISBN: (数字)9798331509828

ISBN: (纸本)9798331509828

Educational institutions frequently suffer from inefficiencies in decision-making, interdepartmental communication, and administrative procedures. In order to optimize resource allocation, automate procurement workflows, and streamline helpdesk operations, this research study proposes an AI-powered platform that combines machine learning (ML) with Natural Language Processing (NLP). The platform acts as a central location for handling enquiries, automating repetitive processes, and enhancing decision-making by utilizing insights derived from data. The proposed system improves query response, fosters interdepartmental communication, and provides predictive analytics to predict resource requirements by continuously learning from user interactions. Through process automation, real-time support, and smooth departmental communication, this AI-driven system aims to improve institutional efficiency, reduce delays, and raise employee satisfaction. © 2025 IEEE.

关键词： Efficiency

来源：评论

学校读者我要写书评

暂无评论

Optimizing Market Interventions a Data-Driven Approach to Price Stabilization

Optimizing Market Interventions a Data-Driven Approach to Pr...

引用

2025 International Conference on Multi-Agent systems for Collaborative Intelligence, ICMSCI 2025

作者： Venkatesh, P. Rupak, S. Robert, Keyas Ebenezar Shakeel Ahamed, E. K.S.Rangasamy College of Technology Department of Computer Science and Business Systems Namakkal India

ISBN: (纸本)9798331509828

This research study proposes data-driven approaches to track and maintain prices of food products. It develops an all-inclusive database of market data based on real-time pricing information generated from reporting centers spread all over the country. Based on ARIMA forecasting models and news API, the system analyzes historical data, seasonal trends, and estimates of crop production to prepare proper predictions regarding future trends in pricing. It senses the price fluctuations and takes a call whether to uncheck the freed buffer stocks or wait for the market. These are projections that would enable active steps towards enabling a feasible and sustainable food supply chain through stabilization of prices, food safety, and optimum use of available resources. © 2025 IEEE.

关键词： Food security

来源：评论

学校读者我要写书评

暂无评论

A Social Robot Companion for Individuals Eating Alone 25

A Social Robot Companion for Individuals Eating Alone

引用

30th International Conference on Intelligent User Interfaces Companion, IUI 2025

作者： Soufargi, Selim Niewiadomski, Radoslaw Fong, Hunter Mancini, Maurizio Department of Computer Science Sapienza University of Rome Rome Italy Department of Informatics Bioengineering Robotics and Systems Engineering University of Genoa Genoa Italy

ISBN: (纸本)9798400714092

In this paper, we present the first implementation of a social robot acting as a companion for individuals eating alone. Our system can engage in multimodal interactions with the user during meals. It conducts conversations on various topics, complemented by gestures and other nonverbal behaviors. The system comprises a NAO robot, human activity recognition modules, a speech recognition module, and a dialog management system integrated with an LLM model. In the paper, we present a technical description of the system and the interaction flow, as well as discuss the possible applications and planned validation. © 2025 Copyright held by the owner/author(s).

关键词： Social robots

来源：评论

学校读者我要写书评

暂无评论

Predictive Health Analysis Using ML for Proactive Risk Detection

Predictive Health Analysis Using ML for Proactive Risk Detec...

引用

2025 International Conference on Multi-Agent systems for Collaborative Intelligence, ICMSCI 2025

作者： Karthik, R. Dhanush, B. Jivitesh Kanna, S. Hari Rama Chandran, A. K.S. Rangasamy College of Technology Department of Computer Science and Business Systems Namakkal India

ISBN: (数字)9798331509828

ISBN: (纸本)9798331509828

The advancements in modern computing technologies have significantly contributed to the development of advanced healthcare monitoring systems., enabling the early detection of critical conditions., such as falls. This review highlights key challenges and applications within healthcare monitoring and prediction models that consider relevant health parameters. In particular., this study discusses about the Random Forest algorithm., which is known for its capacity to manage complex datasets and generate precise patient health forecasts. By optimizing computational efficiency while enhancing predictive accuracy., this study demonstrates the efficacy of machine learning in healthcare. © 2025 IEEE.

关键词： Electronic health record

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：