检索结果-内蒙古大学图书馆

Shadow learner system: implementation of CNN with explainable AI model for bone radiology image classification

Soft Computing 2025年第7期29卷 3571-3588页

作者： Wu, Yaoyang Fong, Simon Liu, Liansheng Department of Computer and Information Science University of Macau Taipa Macao Medical Imaging Department The Sixth Affiliated Hospital of Jinan University Guangdong China Faculty of Digital Science and Technology Macau Millennium College Macao

Advancements in deep learning models for medical image classification have greatly improved accuracy, yet the challenge of explainability persists, making it difficult for clinicians to trust and effectively utilize these models in decision-making. This paper presents the Shadow Learner System, a novel approach that combines Convolutional Neural Networks with Explainable AI (XAI) techniques, focusing on the value of misclassified data—both false positives and negatives—by analyzing them to uncover important insights that traditional methods might overlook. We call our framework the Duplex computer-Aided Diagnosis System. Using the Musculoskeletal Radiographs dataset from Stanford University, which includes images of bone abnormalities that may require surgical intervention, we demonstrate how our system enhances explainability. Our findings show that about 85% of the Regions of Interest identified from misclassified images provided useful insights that clinicians found valuable in their work. Moreover, we observed that the shadow set—comprising important misclassified instances—constituted approximately 4–6% of the original dataset. This research highlights how misclassified data can reveal hidden patterns that improve the overall understanding and trust in AI models used in healthcare. The Shadow Learner System not only serves as a reference for treatment decisions but also has potential applications for other medical conditions that require surgical evaluation, such as osteonecrosis and various tumors. By shifting the focus of XAI from mere model validation to enhancing clinical interpretability, this work contributes to the more effective integration of AI in medical practice. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2025.

关键词： Convolutional neural network Deep learning Explainable AI Medical image classification Regions of interest Shadow learner system

来源：评论

学校读者我要写书评

暂无评论

Security Risks and Best Practices for Blockchain and Smart Contracts: A Systematic Literature Review 8

Security Risks and Best Practices for Blockchain and Smart C...

引用

8th International Conference on information Management and technology, ICIMTech 2023

作者： Yulianto, Semi Hendric Spits Warnars, Harco Leslie Prabowo, Harjanto Meyliana Hidayanto, Achmad Nizar Bina Nusantara University Binus Graduate Program - Doctor of Computer Science Computer Science Department Jakarta11480 Indonesia Binus Business School Undergraduate Program Bina Nusantara University Management Department Jakarta11480 Indonesia School of Information Systems Bina Nusantara University Information Systems Department Jakarta11480 Indonesia Universitas Indonesia Faculty of Computer Science Depok Indonesia

ISBN: (纸本)9798350326093

The rise of blockchain technology and smart contracts has brought widespread attention due to their capacity to transform multiple industrial sectors through decentralized, transparent, secure transactions. However, despite their promise to revolutionize various fields worldwide, lingering concerns regarding security risks impede their adoption rate. Addressing these concerns is crucial now more than ever;therefore, we conducted a comprehensive literature review within our study's scope that focused on published papers between 2014-2023 centered around security risks concerning blockchain and smart contracts. Our systematic approach using the PRISMA checklist analyzed nine categorized research model-based primary studies while recognizing vulnerabilities in smart contract development and providing best practices to mitigate such issues. These findings benefit both researchers and practitioners as they showcase how acknowledging these vulnerabilities can further develop into exploring more significant aspects of blockchain technology's security issues and smart contract development processes. Our study contributes significantly by expanding knowledge in this field while providing novel insights valuable for individuals involved in designing or implementing blockchain technologies. © 2023 IEEE.

关键词： Smart contract

来源：评论

学校读者我要写书评

暂无评论

Identifying The Types and Platforms of the Metaverse - a Systematic Literature Review

Identifying The Types and Platforms of the Metaverse - a Sys...

引用

2022 IEEE Creative Communication and Innovative technology, ICCIT 2022

作者： Suryodiningrat, Satrio Pradono Prabowo, Harjanto Ramadhan, Arief Santoso, Harry Budi Bina Nusantara University Computer Science Binus Graduate Program Jakarta Indonesia Bina Nusantara University Computer Science Graduate Program Jakarta Indonesia Universitas Indonesia Faculty of Computer Science Depok Indonesia

ISBN: (纸本)9781665468763

Metaverse is gaining a lot of traction lately, especially after Mark Zuckerberg's Facebook changed its name to Meta as a preparation to enter another universe called metaverse. The eXtended Reality (XR) technology is also rapidly growing since the birth of Virtual Reality (VR), followed by Augmented Reality (AR), and Mixed Reality is the latest development of XR technology. As the XR technology is the perfect companion to support the metaverse and the understanding of the metaverse is already clear but unfortunately, there is a lot of misconception about the types of the metaverse as well as the platform for each type. This paper aims to point out the types and platforms of the metaverse. 49 articles have been collected and reviewed carefully as a quality assessment process, and 45 articles have been selected for further review. This review informs everybody that there are many types and platforms of the metaverse nowadays. © 2022 IEEE.

关键词： Mixed reality

来源：评论

学校读者我要写书评

暂无评论

Optimal Deep Canonically Correlated Autoencoder-Enabled Prediction Model for Customer Churn Prediction

引用

computers, Materials & Continua 2022年第11期73卷 3757-3769页

作者： Olfat M.Mirza GJose Moses R.Rajender E.Laxmi Lydia Seifedine Kadry Cheadchai Me-Ead Orawit Thinnukool Department of Computer Science College of Computers and Information SystemsUmm Al-Qura UniversityMakkahSaudi Arabia Department of Computer Science Engineering Malla Reddy Engineering College500100India Department of Computer Science Engineering Lendi Institute of Engineering&TechnologyDenkada535005India Department of Computer Science and Engineering Vignan’s Institute of Information TechnologyVisakhapatnam530049India Department of Applied Data Science Noroff University CollegeNorway Program in Statistics and Information Management Faculty of ScienceMaejo UniversityChiang Mai50290Thailand Department of Modern Management and Information Technology College of Arts Media and TechnologyChiang Mai UniversityChiang Mai50200Thailand

Presently,customer retention is essential for reducing customer churn in telecommunication *** churn prediction(CCP)is important to predict the possibility of customer retention in the quality of *** risks of customer churn also get essential,the rise of machine learning(ML)models can be employed to investigate the characteristics of customer ***,deep learning(DL)models help in prediction of the customer behavior based characteristic *** the DL models necessitate hyperparameter modelling and effort,the process is difficult for research communities and business *** this view,this study designs an optimal deep canonically correlated autoencoder based prediction(ODCCAEP)model for competitive customer dependent application *** addition,the O-DCCAEP method purposes for determining the churning nature of the *** O-DCCAEP technique encompasses preprocessing,classification,and hyperparameter ***,the DCCAE model is employed to classify the churners or ***,the hyperparameter optimization of the DCCAE technique occurs utilizing the deer hunting optimization algorithm(DHOA).The experimental evaluation of the O-DCCAEP technique is carried out against an own dataset and the outcomes highlighted the betterment of the presented O-DCCAEP approach on existing approaches.

关键词： Churn prediction customer retention deep learning machine learning archimedes optimization algorithm

来源：评论

学校读者我要写书评

暂无评论

Enhancing Healthcare Data Security: Computational Efficiency of Blockchain-Based Digital Signatures 5th

Enhancing Healthcare Data Security: Computational Efficiency...

引用

5th Doctoral Symposium on Computational Intelligence, DoSCI 2024

作者： Xu, Li Goyal, S.B. Rajawat, Anand Singh Niranjala, T. H. A. Shyama Heshini City Graduate School City University Petaling Jaya46100 Malaysia Nanchang Institute of Science and Technology Nanchang China Faculty of Information Technology City University Petaling Jaya46100 Malaysia School of Computer Science and Engineering Sandip University Nashik India

ISBN: (纸本)9789819771776

This study aims to examine the computational efficiency of blockchain-based digital signature technologies with respect to healthcare data security. When thinking about the revolutionary opportunities and risks posed by healthcare data digitization, concerns regarding data privacy and security have come to the forefront. Blockchain technology, which is both decentralised and immutable, offers a promising alternative. Two of its primary uses, data integrity and identity authentication, are made possible by digital signature systems. We conducted a comprehensive analysis to find out how efficient different protocols are in healthcare computing. Among these protocols were EdDSA, ECDSA, and Schnorr signatures. Our research indicates that efficiency is significantly affected by protocol selection. Despite the strong security it provides, data validation delays may occur due to the increased processing expense generated by ECDSA. Schnorr signatures and EdDSA are viable solutions that offer a reasonable compromise between security and efficiency. By delving into optimization methodologies and associated difficulties, this research paved the road for the ethical application of blockchain technology in healthcare. The healthcare industry, policymakers, and IT developers all benefit from our study when it comes to selecting and improving methods for secure and efficient data storage. This study uniquely examines the computational efficiency of blockchain-based digital signature protocols in healthcare data security, addressing optimization methodologies and highlighting the suitability of Schnorr signatures and EdDSA. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Electronic health record

来源：评论

学校读者我要写书评

暂无评论

Impact of Remote Working During Covid-19 Pandemic on Scrum Team: Experts View on Indonesian E-Commerce Companies Case 2022

Impact of Remote Working During Covid-19 Pandemic on Scrum T...

引用

5th International Conference on Software Engineering and information Management, ICSIM 2022

作者： Nuraga Sani, Kretawiweka Budiardjo, Eko Mahatma, Kodrat Master of Information Technology Faculty of Computer Science Universitas Indonesia Indonesia Faculty of Computer Science Universitas Indonesia Indonesia

ISBN: (纸本)9781450395519

The Covid-19 pandemic has changed many things in our activities, including how we work. Almost all countries, including Indonesia, require remote working or working from home, including technology companies. Many technology companies use Scrum as a framework for their software development. Because of Scrum framework flexibility, organizations with different work cultures can adapt. However, the Covid-19 pandemic forced the team to work remotely and adjust to this separation situation. In this research, we want to determine how this situation influences software development activities in Scrum-based software development in Indonesia. We interview Scrum Experts in three leading e-commerce companies in Indonesia during the pandemic in 2021. Experts found no significant obstacles or impacts when running Scrum during the Covid-19 pandemic remote working. As the research result, the organization needs to consider technical and non-technical aspects to maintain the Scrum team's productivity during the Covid-19 pandemic. Based on interview and survey, as many as 37% of expert Scrum Team state that the product backlog management needs attention because most Scrum teams don't regularly manage the tasks on the product backlog item, so old tasks are drowned out with lots of new tasks added to the list. The Scrum team must also pay more attention to self-discipline to stay focused while working from home. As many as 73% of expert Scrum Team express that the Indonesian E-Commerce Companies must also provide the Scrum team with self-development initiatives to remain motivated. © 2022 ACM.

关键词： Electronic commerce

来源：评论

学校读者我要写书评

暂无评论

Current and Voltage Calculations of the 400 kV Transmission Line Using Laboratory Measurements and computer Simulation 33rd

Current and Voltage Calculations of the 400 kV Transmission...

引用

33rd International Conference on Organization and technology of Maintenance, OTO 2024

作者： Torma, Matej Fekete, Krešimir Kljajić, Ružica Čvek, Kristijan Faculty of Electrical Engineering Computer Science and Information Technology Osijek University of Osijek Kneza Trpimira 2b Osijek31000 Croatia

ISBN: (纸本)9783031805967

In this paper, the current, voltages, as well as active and reactive power of the 400 kV transmission line are calculated using three different methods. The objective of the paper is to demonstrate and compare three different approaches that are based on laboratory measurement, analytical approach, and computer simulation. Measurements are carried out on the laboratory model of a real-life 400 kV transmission line while analytical approach is conducted using a short line model with lumped parameters. As a last approach, the transmission line is modeled in a computer program for power flow analysis, and power flow simulations with the same parameters as for the previous two methods are performed. All three methods are used to calculate voltages, currents, and powers at the beginning and at the end of the transmission line for a few scenarios. Results are compared and discussed in order to demonstrate the possibilities of the applied approaches for the education of power system engineers. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Power transmission lines

来源：评论

学校读者我要写书评

暂无评论

Effect of Introducing Leucine Residues into 2-Aminoisobutyric Acid-Based Amphiphilic Helical Peptides on Intermolecular Interactions and Peptide Self-Assembly

引用

ACS Omega 2025年第21期10卷 22184-22190页

作者： Yamaguchi, Kasumi Kusumoto, Tomoichiro Araki, Takeshi Gomibuchi, Yuki Yasunaga, Takuo Osada, Satoshi Taira, Junichi Department of Chemistry and Applied Chemistry Faculty of Science and Engineering Saga University Saga 840-8502 Japan Department of Bioscience and Bioinformatics Fukuoka 820-8502 Japan Department of Physics and Information Technology Graduate School of Computer Science and Systems Engineering Kyushu Institute of Technology Fukuoka 820-8502 Japan

In recent years, self-assembling de novo-designed peptides have attracted increasing attention as materials for constructing submicrometer-scale structures. We have previously conducted structure-activity relationship studies on the BXBA-20 series of 20-residue amphiphilic peptides containing 2-aminoisobutyric acid (Aib) [Higashimoto, Y. et al. (1999) J. Biochem. 125, 705-712;Hara, T. et al. (2001) J. Biochem. 130, 749-755;Taira, J. et al. (2010) J. Pept. Sci. 16, 607-612]. These peptides share the common sequence Ac-(Aib-Xxx-Aib-Ala)₅-NH₂, where the introduction of a hydrophilic amino acid at the Xxx position results in the formation of a hydrophilic face on one side of the helix. Among them, some peptides exhibited weak intermolecular association and interactions with lipid membranes, though none demonstrated the ability to form submicrometer-scale structures through self-assembly. In this study, we present BKBL-20, a new member of this peptide series incorporating leucine on the hydrophobic face and lysine on the hydrophilic face. Circular dichroism spectroscopy indicated strong association in buffer, while hemolysis assays revealed significant membrane disruption. Transmission electron microscopy further confirmed the formation of ordered fibrous or array-like structures. These findings suggest that BKBL-20 successfully combines self-assembly and membrane-disruptive functions. © 2025 The Authors. Published by American Chemical Society.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Analyzing the Effect of Observer Node Addition Strategy on Bitcoin Double-Spending Attack Detection Using Graph Neural Network 24

Analyzing the Effect of Observer Node Addition Strategy on B...

引用

24th Asia-Pacific Network Operations and Management Symposium, APNOMS 2023

作者： Kang, Changhoon Woo, Jongsoo Hong, James Won-Ki Department of Computer Science and Engineering POSTECH Korea Republic of Graduate School of Information Technology POSTECH Korea Republic of

ISBN: (纸本)9788995004395

Bitcoin transactions are created through the concept called Unspent Transaction Output (UTXO). Users put their own UTXOs as inputs into a transaction for Bitcoin transfer and create multiple outputs, each specifying the recipient’s wallet address and the amount to be sent. UTXO refers to an output that has not been used as an input for any transaction yet and each UTXO can only be used as an input once. However, attempting to use a UTXO more than once is called a double-spending attack. Although double-spending in Bitcoin is ultimately impossible due to the system structure, it can occur when a transaction is deemed confirmed and off-chain goods or services are provided before sufficient transaction finality is guaranteed. We consider an attempt of double-spending attack when a UTXO used as an input in a transaction for payment exists together with another transaction on the Bitcoin network that uses the same UTXO as an input. In previous research, we randomly deployed observer nodes on the Bitcoin network and proposed a method to detect double-spending attacks using transaction data in the memory pool and a graph neural network model. In this paper, we analyze the impact of adding observer nodes to the Bitcoin network on the performance of graph neural network-based Bitcoin double-spending attack detection. We conducted experiments to examine the performance differences among three strategies for adding observer nodes. However, it was difficult to compare clear differences due to the performance degradation of the model caused by the differences in graph structure between datasets. Therefore, we provide an analysis of the causes and suggestions for improvement. Copyright 2023 KICS.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

An Extended Euclidean Algorithm Implementation over Ciphertext Space 2nd

An Extended Euclidean Algorithm Implementation over Ciphert...

引用

2nd International Conference on Security and information Technologies with AI, Internet Computing and Big-data Applications, SITAIBA 2023

作者： Chi, Po-Wen Lin, Yuan-Pang Wang, Ming-Hung Chuang, Yun-Hsin Department of Computer Science and Information Engineering National Taiwan Normal University Taiwan Taipei Taiwan Department of Computer Science and Information Engineering National Chung Cheng University Taiwan Chiayi Taiwan Engineering Management Graduate Program College of Engineering National Cheng Kung University Taiwan Tainan Taiwan

ISBN: (纸本)9789819777853

Fully homomorphic encryption (FHE) is a useful cryptographic primitive. It makes a third-party manipulate data in the ciphertext space and therefore is a key technology for privacy-preserving outsourcing services. Theoretically, FHE can evaluate any circuits. However, the implementation is not trivial. The reason it lacks lots of basic operations like conditional check and branch control. In this paper, we focus on extended Euclidean algorithm operation over FHE. We introduce the problems we encounter and show how we solve these issues. We implement our design and evaluate its performance on a real computer. Finally, we propose some research directions for our future work. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Ciphertext

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：