检索结果-内蒙古大学图书馆

Research Square 2025年

作者： Hamad, Dana Rasul Rashid, Tarik A. Computer Science Department Faculty of Science Soran University Erbil SoranKRG Iraq Computer Science and Engineering Department AIIC University of Kurdistan Hewler KRG Erbil Iraq

Cancer diseases are remaining a leading cause of death in all over the world, this is encouraging the development of advanced diagnostic systems for reliable and accurate detection. The current study investigates the models of traditional machine learning and nature-inspired algorithms for classifying liver, breast and lung cancer cases utilizing structured datasets. More focuses of this research will be on the performance evaluation of five nature-inspired optimizer algorithms like Learner Performance-Based Behavior with Simulated Annealing (LPBSA), Learner Performance-Based Behavior (LPB), FOX Algorithm (FOX), Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Support Vector Machine (SVM), Random Forest (RF), Decision Tree (DT). Among these models LPBSA performing the best, consistently obtained the highest accuracy of 92%, sensitivity 94% and specificity 90%, with confusion matrices reflecting minimal false negative and positives. This work contributes to growing adoption of popular algorithms for enhancing healthcare diagnosis and LPBSA can be utilized in other field investigations. © 2025, CC BY.

关键词： Lung cancer

来源：评论

学校读者我要写书评

暂无评论

Visualization of Community Safety Threats with Time Series Evaluation 5th

Visualization of Community Safety Threats with Time Series E...

引用

5th International Conference on Data science, Machine Learning and Applications, ICDSMLA 2023

作者： Alam, Intekhab Gobi, N. Yadav, Renu Shankar, J. Maharishi School of Engineering and Technology Maharishi University of Information Technology Uttar Pradesh Lucknow India Karnataka Bangalore India Department of Electronic and Communication Engineering School of Engineering and Computing Dev Bhoomi Uttarakhand University Dehradun India School of Computer Science and Engineering Presidency University Karnataka Bengaluru India

ISBN: (纸本)9789819780426

The fast proliferation of connected gadgets on networks and the more and more pervasive use of technology in everyday lives is primary to more danger of cyber threats, making it vital to recognize the behavior of malicious actors as an excellent way to shield each structure and users better. Time collection analysis is a powerful tool to correctly understand and visualize the dynamic conduct of community security threats. This technique can help detect threats as they manifest within the community environment and provide insights into the underlying patterns and tendencies of malicious sports. This era allows us to appropriately come across, classify, and tune safety threats and anomalous community behavior in real-time. With the aid of visualizing the temporal styles of community protection threats, directors can benefit from a green and comprehensive view of the threat panorama and reply quickly. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： anomalous activities directors threats visualizing

来源：评论

学校读者我要写书评

暂无评论

Liver Disease Classification by Analysing Salient Features of Liver Function Test (LFT): An Explainable Machine Learning Approach

Liver Disease Classification by Analysing Salient Features o...

引用

Interdisciplinary Approaches in Technology and Management for Social Innovation (IATMSI), IEEE Conference on

作者： Anmol Agrawal Tirtharaj Sen Amar Sinha Pranabes Gangopadhyay Department of Electronics and Communication Engineering IIIT Naya Raipur India Data Science Virtusa Consulting Services Pvt. Ltd. Hyderabad India Department of Computer Science and Engineering IIIT Naya Raipur India Department of Computer Science and Engineering Asansol Engineering College India

ISBN: (数字)9798331521691

ISBN: (纸本)9798331521707

The liver is a vital organ that performs numerous crucial functions for overall health. However, liver malfunctions can prove fatal, underscoring the urgency of early disease detection. Traditional liver disease classification methods are hampered by complexity and variability, prompting the adoption of Machine Learning (ML) algorithms. In this study, we endeavor to develop a soft computing model leveraging Liver Function Tests (LFTs) for accurate disease classification. Notably, our analysis revealed L2 regularized XGBoost as the standout performer, achieving an impressive 99.94% accuracy. Moreover, our investigation delved deeper into the root causes of liver diseases, shedding light on their underlying mechanisms and relationship to liver function components. By integrating Explainable AI (XAI), our model ensures transparency and provides invaluable insights for informed decision-making. Our project aims to deliver an efficient and robust classification model for liver diseases, poised to significantly impact early detection and patient outcomes.

关键词： Technological innovation Accuracy Machine learning algorithms Liver diseases Explainable AI Computational modeling Decision making Complexity theory Classification algorithms Diseases

来源：评论

学校读者我要写书评

暂无评论

Adaptive Multiscale Attention Networks (ASMA-Net) for Enhanced Diabetic Retinopathy

Adaptive Multiscale Attention Networks (ASMA-Net) for Enhanc...

引用

AI-Driven Smart Healthcare for Society 5.0

作者： Biswadip Basu Mallik Sudeshna Das Anshit Mukherjee Sohini Banerjee Avishek Gupta Department of Basic Science and Humanities Institute of Engineering & Management University of Engineering and Management Kolkata India Department of Computer Science and Engineering Abacus Institute of Engineering and Management Mogra India

ISBN: (数字)9798331536336

ISBN: (纸本)9798331536343

We all know that diabetic retinopathy (DR) is one of the major factors behind vision impairment across the globe and thus creating the demand for accurate and precise detection methods. The traditional detection approaches often depend upon standard convolutional neural networks (CNNs) that frequently struggle to detect complex multi-scale features and contextual data from retinal visualizations. In addition to it, the existing models typically address multi-scale feature extraction and attention strategies independently thus limiting their effectiveness. This work introduces AMSA-Net a novel algorithm that integrates adaptive multi-scale convolutional layers with advanced attention mechanisms. AMSA-Net automatically adjusts the scale of attention provided to various regions of retinal images and thus enhancing the model’s capability to capture both local and global characteristics important for diagnosing DR at various stages. According to scientific conclusion, our AMSA-net surpass other advanced strategies of this domain of DR detection and thus achieving higher accuracy and robustness across diverse datasets and this we confirmed through simulating tools. Our model effectively identifies DR symptoms by highlighting relevant regions of interest while suppressing irrelevant details thus improving the extraction of features. By highlighting the limitations of current DR detection strategies, AMSA-Net actually offers a reliable solution for automated screening in ophthalmology. The enhanced flexibility and accuracy of AMSA-Net could at times result in detection and enhanced benefit yields for the patients, and hence opening the door towards better intervention in control of diabetic retinopathy.

关键词： Training Diabetic retinopathy Adaptation models Accuracy Attention mechanisms Web services Visual impairment Feature extraction Retina Inference algorithms

来源：评论

学校读者我要写书评

暂无评论

Detecting and Characterising Mobile App Metamorphosis in Google Play Store

引用

IEEE Transactions on Mobile Computing 2025年

作者： Denipitiyage, D. Silva, B. Gunathilaka, K. Seneviratne, S. Mahanti, A. Seneviratne, A. Chawla, S. University of Sydney School of Computer science Australia University of Moratuwa Department of Computer Science and Engineering Sri Lanka Sydney Australia Qatar Computing Research Institute Qatar

App markets have evolved into highly competitive and dynamic environments for developers. While the traditional app life cycle involves incremental updates for feature enhancements and issue resolution, some apps deviate from this norm by undergoing significant transformations in their use cases or market positioning. We define this previously unstudied phenomenon as 'app metamorphosis.' In this paper, we propose a novel and efficient multi-modal search methodology to identify apps undergoing metamorphosis and apply it to analyse two snapshots of the Google Play Store taken five years apart. Our methodology uncovers various metamorphosis scenarios, including re-births, re-branding, re-purposing, and others, enabling comprehensive characterisation. Although these transformations may register as successful for app developers based on our defined success score metric (e.g., re-branded apps performing approximately 11.3% better than an average top app), we shed light on the concealed security and privacy risks that lurk within, potentially impacting even tech-savvy end-users. © 2002-2012 IEEE.

关键词： Mobile commerce

来源：评论

学校读者我要写书评

暂无评论

A Deep Neural Framework for Self-Injurious Behavior Detection in Autistic Children

引用

Procedia computer science 2025年 258卷 3490-3499页

作者： Uday Singh Shailendra Shukla Manoj Madhava Gore Computer Science and Engineering Department Motilal Nehru National Institute of Technology Allahabad Prayagraj Utter Pradesh India

Video Action Detection (VAD) is becoming increasingly common, with distributed methods shifting towards edge computing for real-time processing. The limited diversity and size of the existing Self-Stimulatory Behaviours Dataset (SSBD) hinder the gen-eralizability of Self-Injurious Behaviors (SIB) detection models. Early detection of SIB is imperative for timely intervention and support for individuals with Autism Spectrum Disorder (ASD), emphasizing the need for accurate recognition systems. Addressing these challenges requires advancements in dataset diversity, model accuracy, and generalizability to diverse populations and environments. This paper proposes a framework for the detection of SIB in children. We use a hybrid approach combining CNN and LSTM to detect SIB effectively. To enhance the recognition capabilities, we add new actions to the SSBD dataset to increase its size. This addition enriches the dataset, enabling more comprehensive training of our recognition models. We then extract frames from videos and apply augmentation techniques to the frames. The ConvLSTM, EfficientNet, and Long-Short Term Recurrent Convolutional Networks (LRCN) models are used for SIB action detection. Among these, the LRCN model demonstrates superior performance, achieving an accuracy of 92.62%, surpassing ConvLSTM (80.33%) and EfficientNet (77.17%). The LRCN model achieved a Mean Squared Error (MSE) of 0.045, highlighting its reliability in minimizing prediction errors for action detection. This underscores the effectiveness of hybrid models for video action recognition, emphasizing the importance of early detection in supporting individuals with ASD.

关键词： Machine Learning Autism Self Injurious Behaviour video classification Deep Learning repetitive behavior

来源：评论

学校读者我要写书评

暂无评论

An Advanced Cognitive Approach for Personalized Drug Formulation in Type 2 Diabetes Using Deep Neural Networks and Feature-based Fusion

An Advanced Cognitive Approach for Personalized Drug Formula...

引用

2025 International Conference on Intelligent Control, Computing and Communications, IC3 2025

作者： Singh, Brijendra Pratap Singh, Jagendra Suryavanshi, Madhuri Jailani, S. Amer, A. Beer A. Kumar, Rakesh School of Computer Science Engineering & Technology Bennett University Greater Noida India Pimpri Chinchwad College of Engineering Pune India Alpha Pharma Kaec R&d Formulation Dev. Rabigh Saudi Arabia Sadat academy for management and Sciences Canob street Alexandria Egypt Magadh University Department of Computer Science and I.T. Bodh Gaya India

ISBN: (纸本)9798331527495

Personalized medicine is an innovative approach toward achieving improved therapeutic outcomes, notably in chronic diseases like Type 2 Diabetes (T2D), where individualised drug formulations represent the crux of treatment. Conventional methods of drug prescription tend to focus on generic approaches without considering the specific genetic, physiological, and lifestyle characteristics of individuals. This research fills this gap by developing an advanced deep learning-based method for personalised drug formulation in T2D. It integrates diverse patient data, including demographics, medical history, clinical measurements, genomic information, pharmacokinetic information, real-time monitoring data, and responses to previous treatments. For optimal drug formulation, it makes use of a combination of deep learning models, such as Recurrent Neural Network (RNN) for time-series prediction, Convolutional Neural Network (CNN) for genomic analysis, GAN for producing synthetic data, and auto-encoder for feature extraction. The dataset consists of 2700 patient records collected over one year, wherein the pre-processing phase includes handling missing values, normalising continuous variables, and encoding categorical features. Trained models were assessed based on precision, recall, F1-score, and AUC ROC levels;the best performance came from the RNN model with a maximum accuracy of 97.35%. CNN, GAN, and Autoencoder also obtained similar competitive results. However, the RNN-based model received the highest score in recommending the appropriate drugs. The application of this research can contribute to the broader discipline of precision medicine. © 2025 IEEE.

关键词： Theranostics

来源：评论

学校读者我要写书评

暂无评论

Enhancing Healthcare with Blockchain: Innovations in Data Privacy, Security, and Interoperability

Enhancing Healthcare with Blockchain: Innovations in Data Pr...

引用

Disruptive Technologies (ICDT), International Conference on

作者： Bhupendra Kumar Vishal Garg Kahksha Ahmed Puneet Garg Shweta Choudhary Pashupati Baniya Department of Computer Science & Engineering ITS Engineering College (AKTU) Greater Noida India NCOG Limited Corporation Faridabad India Computer Science and Engineering St. Andrews Institute of Technology and Management Gurugram Delhi NCR India Department of Computer Science & Engineering (AI) KIET Group of Institutions Delhi NCR Ghaziabad India Applied Science Department ABES Engineering College Ghaziabad India Computer Science and Engineering IEC College of Engineering and Technology (AKTU) Greater Noida India

ISBN: (数字)9798331519582

ISBN: (纸本)9798331519599

The rise of digital technologies has transformed the healthcare industry, enabling smarter and more efficient services. But high data volumes and reliance on centralized systems pose significant challenges, including real-time data sharing functionality and vulnerability to breaches. Blockchain technology offers flexible solutions for distributing data management and ensuring secure, transparent and immutable transactions. This paper explores the use of blockchain in healthcare, emphasizing its potential to improve data privacy, improve communication, and simplify the management of electronic health records (EHRs). Key features of the blockchain such as decentralization, immutability and cryptographic security have been analyzed in the context of healthcare. The review looks at notable developments, including blockchain-based frameworks for secure data shares, smart contracts for automated processes, and parallel healthcare systems combining it with artificial intelligence, as well as enabling efficiencies within the supply chain. This work provides a comprehensive analysis of blockchain's transformative potential and identifies areas for further research to ensure that it has been widely accepted.

关键词： Data privacy Technological innovation Supply chains Smart contracts Medical services Transforms Solids Blockchains Stakeholders Artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

Comprehensive Review on Sales Prediction Models

引用

Procedia computer science 2025年 259卷 1218-1227页

作者： Prabu C R Rahulkrishnan Ravindran Katari Sai Abhiram Varma Kantipudi Sai Sri Rohith Gannamaneni Sowmya Remya M S Department of Computer Science and Engineering Amrita School of Computing Amrita Vishwa Vidyapeetham Amritapuri 690525 Kerala India

This study presents a comprehensive investigation into the performance of multiple machine learning algorithms, emphasizing its fundamental aim to conduct a comparative study of diverse machine learning algorithms. Subsequently, this paper delves into the complexities and essential features, setting the stage for the subsequent analytical exploration. The Rossmann Sales dataset is used as a baseline for analytic comparison. Using Light Gradient Boosting Machine, Elastic net regression, XG Boost, Ransac Regression, Linear Regression, Ridge Regression, Random Forest, Decision Tree, K-Nearest Neighbors, and Gradient Boosting, the research focuses on its complexities and key components. Along with evaluating each model’s performance using metrics like MAE, RMSE, and R-squared, the study also looks at the theoretical underpinnings and practical mechanisms of the algorithms. Informed decision-making in the sales analytics and predictive modeling industries is supported by the results, which shed light on the viability and applicability of different models for forecasting Rossmann store sales.

关键词： Light Gradient Boosting Machine Elastic net regression XG Boost Ransac Regression Linear Regression Ridge Regression Random Forest DecisionTree K-Nearest Neighbors Gradient Boosting Mean Absolute Error Root Mean Squared Error

来源：评论

学校读者我要写书评

暂无评论

Ensuring Digital Sovereignty in Cross-chain EHR Sharing: A Relay-as-a-Service Approach for Secure Healthcare Interoperability

引用

Procedia computer science 2025年 254卷 48-57页

作者： Dharavath Ramesh Thakur Santosh Munesh Chandra Trivedi Chi Hieu Le Department of Computer Science and Engineering Indian Institute of Technology (Indian School of Mines) Dhanbad Jharkhand India Department Centre for Life Sciences Ecole Centrale School of Engineering Mahindra University Hyderabad India Department of Computer Science and Engineering National Institute of TechnologyAgartala Tripura India Faculty of Engineering and Science University of Greenwich Kent ME4 4TB UK

Electronic Health Records (EHRs) stored in cloud environments often face privacy challenges in healthcare data management due to the divide between patient ownership and institutional control. Blockchain technology offers a promising solution with its features of immutability and traceability. However, existing blockchain-based approaches for EHR privacy preservation are limited to single institutions and fail to address the critical need for cross-chain compatibility and digital sovereignty. To bridge this gap, we propose two novel strategies: Polkadot-based Cross-chain for EHR-preserving Blockchain (PCEB) and Relay-as-a-Service-based Cross-chain for EHR-preserving Blockchain (RaSCEB). PCEB utilizes Polkadot's relay communication to securely share EHR data across multiple healthcare networks while preserving patient privacy and ensuring digital sovereignty. RaSCEB introduces Relay-as-a-Service (RaaS) to enable seamless EHR sharing across blockchain ecosystems, empowering patients with control over their data while maintaining regulatory compliance and sovereignty over their digital health records. Both approaches are validated through comprehensive security analysis and performance evaluations. We also present an interoperability framework tailored for permissioned blockchain networks, emphasizing trust derived from consensus mechanisms. Our work addresses the urgent need for cross-chain compatibility in EHR privacy preservation and advances interoperability solutions while safeguarding digital sovereignty in healthcare and beyond.

关键词： Electronics Health Records (EHRs) Blockchain interoperability Cross blockchain Digital sovereignty

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：