检索结果-内蒙古大学图书馆

Addressing challenges in accurate brain tumor classification in MRI: a transfer learning approach with EfficientNetB3 and comprehensive model evaluation

引用

Multimedia Tools and Applications 2024年 1-31页

作者： Das, Surajit Goswami, Rajat Subhra Department of Computer Science and Engineering National Institute of Technology Arunachal Pradesh Arunachal Pradesh Jote791113 India

Brain tumors pose significant health risks because of uncontrolled and abnormal cell growth, potentially leading to the devastating of certain organs and even death, mostly in adult populations. So, early and precise brain tumor classification (BTC) is essential, as a timely diagnosis directly impacts the efficacy of treatment and the survival rates of patients. Magnetic Resonance Imaging (MRI) is the preferred diagnostic tool for BTC due to its superior image quality, yet manual classification methods have the disadvantage of being time-consuming and labour-intensive. The maturity of deep learning (DL) in modern times has brought remarkable progress in automating complex processes across multiple domains, including transfer learning (TL) in medical imaging. In our study, we propose an approach to partially automate the BTC task through the application of transfer learning on MRI data. After applying image pre-processing techniques like cropping, enhancement and Augmentation, we tested six architectures - EfficientNetB6, EfficientNetB3, VGG19, NASNetLarge, DenseNet169 and DenseNet201. Our experiments on a comprehensive publicly available kaggle dataset show that the EfficientNetB3 model achieved the best performance, with a training accuracy of 100% and test accuracy of 99.62%, outperforming all other architectures. EfficientNetB6 demonstrated a similar accuracy of 99.62%, while DenseNet201 and NASNetLarge performed well, with accuracies of 99.24% and 99.08%, respectively. VGG19, although a popular baseline model, achieved a lower accuracy of 94.96%. These findings suggest that EfficientNetB3, alongside other advanced architectures, offers a robust and accurate solution for brain tumor classification in MRI, paving the way for more reliable and efficient diagnostic tools in clinical settings. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Magnetic resonance imaging

来源：评论

学校读者我要写书评

暂无评论

A Review of Modern Computing Paradigms: Cloud, Fog, and Edge

A Review of Modern Computing Paradigms: Cloud, Fog, and Edge

引用

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

作者： Sharma, Neha Choudhary, Monika Srivastava, Prakash Prabha, Chander Sharma, Himanshu Cgc Department of Computer Science and Engineering Punjab Landran India Department of Computer Science & Engineering Dehradun India Chitkara University Institute of Engineering and Technology Chitkara University Punjab India

ISBN: (纸本)9798331509828

In the current competitive environment, businesses are looking at how to lower their IT expenses while simultaneously improving their productivity and flexibility to respond quickly to changes in their operational procedures. One of the most widely accepted information technology-based services is cloud computing, which has garnered interest from both the academic and business communities as well as the general public-characteristics such as pay-per-use and subscription options ease of use, elasticity, and low entry costs. Low entry cost in fog, cloud, and edge computing is achieved by utilizing pay-as-you-go pricing models, leveraging open-source technologies, and optimizing resource allocation. By emphasizing how this technology differs from the rest of the current technology and concentrating on its numerous benefits and areas of application-which provide proof of its growing popularity-this paper presents the development of modern computer paradigms and associated areas of research interest. A discussion of each paradigm is done by clearly defining its main ideas and how they relate to the others. The paper presents a comparative assessment of Cloud, Edge, and Fog computing indicating their diverse areas of applicability. © 2025 IEEE.

关键词： Fog computing

来源：评论

学校读者我要写书评

暂无评论

Designing Of Hybrid Intrusion Detection System With Combined Approach Of LightGBM-AE And Double DQN 27

Designing Of Hybrid Intrusion Detection System With Combined...

引用

27th International Symposium on Wireless Personal Multimedia Communications, WPMC 2024

作者： Sinha, Ditipriya Srivastava, Arpita Singh, Om National Institute of Technology Patna Department of Computer Science and Engineering Patna India Indian Institute of Engineering Science and Technology Shibpur Department of Information Technology Shibpur India

ISBN: (纸本)9798350392319

Cyber-attacks are a growing threat to individuals, businesses, and governments. Data science helps improve security by analyzing network traffic and system activities. Intrusion detection systems (IDSs) are essential for enhancing security and resilience, but IDS datasets face challenges in obtaining quality training data and accurately predicting cyberattacks. To tackle these issues, this paper designs a novel hybrid intrusion detection system that detects malicious behavior based on signature and anomaly. It utilizes advanced techniques like one hot encoding, min-max normalization for data preprocessing, modified GAMO for data oversampling, information gain, and modified grey wolf optimizer (GWO)based bi-phase approach for feature selection, stacking of LightGBM with autoencoder for signature-based detection, and finally double deep Q-network (Double DQN) for anomaly-based detection to create robust and efficient hybrid IDS model. The proposed model achieves 94.33% and 88.90% accuracy utilizing signature-based and anomaly-based IDS, respectively, outperforming state-of-the-art 'MTH-HIDS' in accuracy, precision, recall, and f1-score. © 2024 IEEE.

关键词： Network intrusion

来源：评论

学校读者我要写书评

暂无评论

A Novel Blockchain-Assisted Deep Learning Model For Secure Edge Intelligence in IoT Networks

引用

Journal of The Institution of Engineers (India): Series C 2024年 1-14页

作者： Sekhar, G. Chandra Aruna, R. Department of Computer Science and Engineering Vel Tech Rangarajan Dr.Sagunthala R & ampD Institute of Science and Technology Tamilnadu Chennai India

The Internet is being connected to billions of mobile and Internet of Things (IoT) devices, generating zillions of Bytes of data at the network edge. This is in addition to developments in mobile computing and the Internet of Things (IoT). In the context of edge computing, applications are placed closer to data sources such as IoT sensors, local edge devices, and edge servers. Data are transmitted from edge devices to dedicated computer infrastructures for a range of activities in any edge computing environment. For the reason that data are regularly sent from IoT edge devices to centralized virtualization technologies, it is more vulnerable to security vulnerabilities and complexity measures. Finally, to preserve the privacy of the IoT devices, the performance of the proposed algorithms is evaluated using datasets in a large-scale network prototyped system. According to the data, the suggested strategy is both more effective and secure than the alternatives. © The Institution of Engineers (India) 2024.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

HSPBCI: a robust framework for secure healthcare data management in blockchain-based IoT systems

引用

Multimedia Tools and Applications 2024年 1-25页

作者： Gupta, Sangeeta Chithaluru, Premkumar Stephan, Thompson Nafisa, Shaik Kumar, Sandeep Department of Computer Science and Engineering Chaitanya Bharathi Institute of Technology Hyderabad500075 India Department of Information Technology Mahatma Gandhi Institute of Technology Telangana Hyderabad500075 India Department of Computer Science and Engineering Graphic Era Deemed to Be University Uttarakhand Dehradun India Centre of Research Impact and Outreach Chitkara University Institute of Engineering and Technology Chitkara University Punjab India

Abstract: Online monitoring and counseling have helped patients and made digital health management possible. However, the vulnerability of medical data gathered via devices, especially in the Internet of Things (IoT) context, highlights the necessity of strong security protocols. IoT is essential, but there are still difficulties in managing vital data over a heterogeneous network. To reduce potential health hazards, it is essential to ensure that IoT sensors do not negatively impact human health. Maintaining the privacy of patient records is crucial to avoid unauthorized access or financial gain by third parties. This paper introduces the Healthcare System using Private Blockchain-based Cloud IoT (HSPBCI) services, providing a comprehensive framework to address these issues. The framework's main goals are to speed up block-based transaction encryption and decryption times and grant authorized access to patient records. The proposed HSPBCI model restricts access to patient records to those holding the appropriate key, ensuring secure block-based access. Nodes are randomly deployed in cloud locations to generate data, which is then processed by the cloud blockchain network to filter out unscrupulous users and service providers. The results show that the proposed HSPBCI model, with up to 50 nodes, has encryption and decryption times that are over five times faster than existing models. With an improvement of a factor of O(n) over the existing O(n^2), the proposed model's transaction processing and runtimes for different characteristics and nodes range from 0.15 ns to 0.44 ns and 1 ms to 470 ms, respectively, compared to 0.18 ns to 0.53 ns and 2 ms to 570 ms in existing works. In the context of blockchain-based IoT applications, this research advances the field of secure healthcare data management. Graphical Abstract: (Figure presented.) © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Electronic health record

来源：评论

学校读者我要写书评

暂无评论

EnsemBEP: A Robust Ensemble Model for Enhanced B-cell Epitope Prediction 8

EnsemBEP: A Robust Ensemble Model for Enhanced B-cell Epitop...

引用

8th International Conference on Electronics, Communication and Aerospace technology, ICECA 2024

作者： Karthiga, M. Shiv Navin, V.B. Cibin Kumar, R.M. Sathiya, B. Satheesh Kumar, A. Priyanga, M.A. Bannari Amman Institute of Technology Department of Computer Science and Engineering Tamilnadu Sathyamangalam India Nandha Engineering College Department of Computer Science and Engineering Tamilnadu Erode India

ISBN: (纸本)9798350367904

B-cell epitope prediction has found its use in understanding B-cell recognition spots on a number of antigens that is vital in vaccine design and immunotherapy. These fields urgently need powerful prediction models, but the existing methods often fail to satisfy the requirement of both high sensitivity and high specificity. In response, this manuscript presents an ensemble model named EnsemBEP that combines Random Forest, Gradient Boosting, and XGBoost for improving the performance of B-cell epitope prediction. New model for epitope prediction, named EnsemBEP shows better performance in comparison with other models including EpiDope and BepiPred-2.0. The proposed EnsembEP model demonstrates superior performance compared to existing models such as EpiDope and BepiPred-2.0. Through rigorous evaluation, EnsemBEP achieved an AUC of 0.945, accuracy of 0.885, sensitivity of 0.823, specificity of 0.926, and an MCC of 0.743. All these findings point out that EnsemBEP yield better results than competing models thereby confirming that it is more accurate and reliable in predicting B-cell epitopes. The results provide evidence for the significance of ensemble learning methodologies in attaining higher predictive performance in immunoinformatic. © 2024 IEEE.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

Enhancing object detection for visually impaired integrating YOLOv8 with spiking EfficientDet

引用

Multimedia Tools and Applications 2025年 1-29页

作者： Biradar, Vinod Gull, Karuna C. Department of Computer Science and Engineering KLE College of Engineering & Technology Karnataka Chikodi591201 India Department of Computer Science and Engineering S. G. Balekundri Institute of Technology Belagavi Visvesvaraya Technological University Karnataka Belagavi590018 India

Object detection is a significant method in computer vision, mainly for detecting objects belonging to different classes in an image. Its applications are widely spread to video surveillance, human tracking, autonomous vehicles, and assistive technologies for sight-impaired individuals. However, existing methods suffer from accurately identifying an object and its respective classes due to misidentifying or failing to identify smaller objects. Therefore, this research proposes a new object detection model called You Only Look Once version 8 with Spiking EfficientDet (Yv8SED) that improves detection accuracy, reducing wrong classification with particular emphasis on smaller objects. This model is beneficial in assistive devices for the navigation and safety of a visually-impaired person. The Yv8SED method provides excellent object detection performance with reduced time and cost, making it an efficient methodology for different object detection tasks. The segmentation process is improved through SegNet, which efficiently separates the object from the image. Moreover, the Hippopotamus Optimization Algorithm (HOA) is used to optimize error parameters in the proposed method, thereby improving efficiency and robustness in the object detection process. Experimental results proved the efficiency of the proposed method, with mAP at 95% and mean Average Recall (mAR) at 93% on the COCO dataset and mAP and mAR at 95% and 98%, respectively, on the VOC dataset. These results verify the efficacy and reliability of the proposed method for object detection, particularly for enhancing assistive technologies for the blind and visually impaired. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Implementation of Secure and Verifiable Access Control Procedures using the NTRU Cryptosystem to Store Big Data in the Cloud Environment

引用

Recent Patents on engineering 2025年第2期19卷 57-70页

作者： Babu B, Ravindra Saxena, Swati Widjaja, Gunawan Vinodha D, Vedha Omarov, Batyrkhan Ramaiah, Gurumurthy B. Vats, Prashant Adama Ethiopia Department of Computer Science KCG College of Technology Chennai India Universitas Krisnadwipayana Jawa Barat 13077 Indonesia Department of Electronics and Communication Engineering JCT College Of Engineering and Technology Pichanur Coimbatore India International University of Tourism and Hospitality Turkistan Kazakhstan Technical and Vocational Training Institute Addis Ababa Ethiopia Department of Computer Science and Engineering Faculty of Engineering and Technology SGT University Haryana Gurugram India

Background: Due to their complexity and size, deploying ciphertexts for clouds is considered the most useful approach to accessing large data stores. Methods: However, access to a user's access legitimacy and improving a decrypted text on the cloud depending on an improved access policy (AP) specified by the data owner are the key challenges for making large data storage realistic and effective in clouds. The traditional ways either totally remove the problem of AP development or offer renewal to arbiter power, but in real-time, enhancing the AP is essential to maximising security and handling agility. Results: In this patent paper, a safe and verifiable access control program characterised by the NTRU cryptographic system for large storage of data in the clouds is proposed. Primarily, an improved NTRU decryption protocol to deal with the decryption failures of the prime NTRU is established, and in addition, the program is analysed for its security strength and computational performance. When a new AP is specified by the data owner, the cloud server allows the program to improve ciphertext and allows the owner to verify the upgrade to oppose the cloud's fraudulent behaviour. Conclusion: It enables (i) checking the user's legitimacy to access the data owner and qualified users, and (ii) allowing the user to verify the data provided by other users for the recovery of the right user. Strong analysis can prevent and block delinquency from various attacks, namely the collusion attack that could potentially target fraud users. © 2025 Bentham science Publishers.

关键词： Cloud security

来源：评论

学校读者我要写书评

暂无评论

Enhancing LLM Performance on Legal Textual Entailment with Few-Shot CoT-based RAG

Enhancing LLM Performance on Legal Textual Entailment with F...

引用

2024 IEEE International Conference on Signal Processing, Informatics, Communication and Energy Systems, SPICES 2024

作者： Reji, Sneha Ann Sheik, Reshma Sharon, A. Avisha Rai, M. Jaya Nirmala, S. National Institute of Technology Department of Computer Science and Engineering Tamil Nadu Tiruchirappalli India TKM College of Engineering Department of Computer Science and Engineering Kerala Kollam India

ISBN: (纸本)9798350376135

Large Language Models (LLMs), leveraging extensive data, have become successful in text generation applications. However, their integration into the legal domain presents distinct challenges, prompting deviations in thought processes and potentially misleading outcomes. Such discrepancies can significantly impact the efficacy of legal text applications. In our study, we propose a novel approach that combines retrieval-based methods with generation-based techniques to provide relevant explanations and labels for the given query for legal textual entailment task, leveraging the concept of Chain of Thought (CoT) for reasoning and few-shot prompting. Later, we enhance LLM outputs through the application of various text generation strategies. Our findings demonstrate an enhancement, with a 5% increase in F1 score and a 3% improvement in accuracy when compared with zero-shot baselines. This highlights the effectiveness of utilizing these strategies in specialized fields such as law, where precision and reliability are of utmost importance. © 2024 IEEE.

关键词： Modeling languages

来源：评论

学校读者我要写书评

暂无评论

A GMM-CGABC Based Energy Replenishment Scheme for On-Demand Wireless Rechargeable Sensor Networks 1

A GMM-CGABC Based Energy Replenishment Scheme for On-Demand ...

引用

1st International Symposium on Parallel Computing and Distributed Systems, PCDS 2024

作者： Mouli, S.Sudhamsu Veeraiah, T. Mahesh, K. Singh, M.P. Ecole Centrale School of Engineering Mahindra University Department of Computer Science and Engineering Hyderabad India National Institute of Technology Department of Computer Science and Engineering Patna India

ISBN: (纸本)9798350349658

Wireless Sensor Networks (WSNs) are constrained by the limited energy capacity of Sensor Nodes (SNs), which hinders their perpetual operation. The advent of Wireless Energy Transfer (WET) technology has emerged as a promising solution to this challenge, enabling the creation of Wireless Rechargeable Sensor Networks (WRSNs). Recent research has shown growing interest in integrating Wireless Mobile Chargers (WMCs) into WRSNs to replenish the energy reserves of critical SNs through WET. A key aspect of WET involves using one or more WMCs to recharge SNs, with optimizing the charging schedule (i.e., the sequence in which SNs are charged) crucial in on-demand WRSNs. In addition, spatiotemporal constraints imposed by the requesting SNs must be considered when devising the WMC's charging schedule. To address these challenges, this work proposes the integration of a Gaussian Mixture Model with a Chaotic Gaussian Artificial Bee Colony (GMM-CGABC) to develop an efficient recharge schedule that revitalizes SNs. This model is designed to optimize SN deployment, maximize network coverage, minimize energy consumption, and extend network longevity. The proposed method determines the optimal charging path for WMCs within each cluster to minimize Average Charging Latency (ACL) while incorporating a novel strategy for fitness function evaluation. Extensive simulations demonstrate that the proposed system outperforms state-of-the-art algorithms, including First Come, First Served (FCFS), Nearest Job Next with Preemption (NJNP), and the Gravitational Search Algorithm (GSA), in terms of charging latency. © 2024 IEEE.

关键词： Inductive power transmission

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：