Blockchain-Based Privacy-Enhancing Federated Learning in Smart Healthcare: A Survey

Blockchains, Journal Year: 2025, Volume and Issue: 3(1), P. 1 - 1

Published: Jan. 1, 2025

Federated learning (FL) has emerged as an efficient machine (ML) method with crucial privacy protection features. It is adapted for training models in Internet of Things (IoT)-related domains, including smart healthcare systems (SHSs), where the introduction IoT devices and technologies can arise various security concerns. However, FL cannot solely address all challenges, privacy-enhancing (PETs) blockchain are often integrated to enhance frameworks within SHSs. The critical questions remain regarding how these they contribute enhancing This survey addresses by investigating recent advancements on combination PETs healthcare. First, this emphasizes integration into context. Second, challenge integrating FL, it examines three main technical dimensions such blockchain-enabled model storage, aggregation, gradient upload frameworks. further explores collectively ensure integrity confidentiality data, highlighting their significance building a trustworthy SHS that safeguards sensitive patient information.

Language: Английский

---

Deep Learning in the Fast Lane: A Survey on Advanced Intrusion Detection Systems for Intelligent Vehicle Networks

IEEE Open Journal of Vehicular Technology, Journal Year: 2024, Volume and Issue: 5, P. 869 - 906

Published: Jan. 1, 2024

The rapid evolution of modern automobiles into intelligent and interconnected entities presents new challenges in cybersecurity, particularly Intrusion Detection Systems (IDS) for In-Vehicle Networks (IVNs). This survey paper offers an in-depth examination advanced machine learning (ML) deep (DL) approaches employed developing sophisticated IDS safeguarding IVNs against potential cyber-attacks. Specifically, we focus on the Controller Area Network (CAN) protocol, which is prevalent in-vehicle communication systems, yet exhibits inherent security vulnerabilities. We propose a novel taxonomy categorizing techniques conventional ML, DL, hybrid models, highlighting their applicability detecting mitigating various cyber threats, including spoofing, eavesdropping, denial-of-service attacks. highlight transition from traditional signature-based to anomaly-based detection methods, emphasizing significant advantages AI-driven identifying intrusions. Our systematic review covers range AI algorithms, neural network such as Transformers, illustrating effectiveness applications within IVNs. Additionally, explore emerging technologies, Federated Learning (FL) Transfer Learning, enhance robustness adaptability solutions. Based our thorough analysis, identify key limitations current methodologies paths future research, focusing integrating real-time data cross-layer measures, collaborative frameworks.

Language: Английский

---

A Comprehensive Review of Deep Learning: Architectures, Recent Advances, and Applications

Information, Journal Year: 2024, Volume and Issue: 15(12), P. 755 - 755

Published: Nov. 27, 2024

Deep learning (DL) has become a core component of modern artificial intelligence (AI), driving significant advancements across diverse fields by facilitating the analysis complex systems, from protein folding in biology to molecular discovery chemistry and particle interactions physics. However, field deep is constantly evolving, with recent innovations both architectures applications. Therefore, this paper provides comprehensive review DL advances, covering evolution applications foundational models like convolutional neural networks (CNNs) Recurrent Neural Networks (RNNs), as well such transformers, generative adversarial (GANs), capsule networks, graph (GNNs). Additionally, discusses novel training techniques, including self-supervised learning, federated reinforcement which further enhance capabilities models. By synthesizing developments identifying current challenges, insights into state art future directions research, offering valuable guidance for researchers industry experts.

Language: Английский

---

Secure internet of medical things based electronic health records scheme in trust decentralized loop federated learning consensus blockchain

International Journal of Intelligent Networks, Journal Year: 2024, Volume and Issue: 5, P. 161 - 174

Published: Jan. 1, 2024

Electronic Health Records (EHRs) have become an increasingly significant source of information for healthcare professionals and researchers. Two technical challenges are addressed: motivating federated learning members to contribute their time effort, ensuring accurate aggregation the global model by centralized server. To overcome these issues establish a decentralized solution, integration blockchain proves effective, offering enhanced security privacy smart healthcare. The proposed approach includes gamified element incentivize recognize contributions from members. This research work offers solution involving resource management within Internet Medical Things (IoMT) using newly trust loop consensus blockchain. obtained raw data is pre-processed handling missing values adaptive min-max normalization. appropriate features selected with aid hybrid weighted-leader exponential distribution optimization algorithm. Because, multiple exhibits varying levels variation across each feature. then forwarded training phase through pyramid squeeze attention generative adversarial networks classify EHR as positive negative. classification demonstrates high flexibility scalability, making it applicable wide range network architectures various computer vision tasks. introduced provides better outcomes in terms 98.5% accuracy 99% validation over Information Mart Intensive Care III (MIMIC-III) dataset, which more efficient than other traditional methods.

Language: Английский

---

Inclusive Role of Internet of (Healthcare) Things in Digital Health: Challenges, Methods, and Future Directions

Published: Jan. 3, 2025

Healthcare systems could undergo a change with the incorporation of Internet Things (IoT) technology, which would allow for enhanced analytics, real-time data monitoring, and seamless communication. This chapter offers thorough analysis IoT applications in healthcare systems, emphasizing their influence on several facets delivery, such as patient care, digital health, preventative medicine, remote future directions IoT-enabled systems. The first section covers basic elements industry, including networks, sensors, linked devices, cloud-based platforms. It looks at how wearable smart devices that continuous health tracking processing, monitoring can all improve care. In context healthcare, potential to support personalized treatment early diagnosis intervention is explored. addition, investigates integration affects medical electronic records, hospital management help organizations make better decisions, use resources more effectively, run operations efficiently. There also discussion difficulties factors involved implementing IoT, privacy, interoperability, security, regulatory compliance. emphasizes have revolutionize way global issues, managing chronic diseases, aging populations, restricted access services places, are addressed. highlights remove obstacles promote widespread stakeholders, technology developers, policymakers, researchers, providers, must work together. Finally, application presents plethora chances boost increase operational effectiveness, spur innovation provision services. however, need be carefully considered. system may adopt patient-centered, data-driven strategy enhance outcomes by utilizing IoT. will delivered age.

Language: Английский

---

Computational intelligence for sustainable computing in health care informatics

Elsevier eBooks, Journal Year: 2025, Volume and Issue: unknown, P. 163 - 180

Published: Jan. 1, 2025

Language: Английский

---

A Novel Framework for Quantum-Enhanced Federated Learning with Edge Computing for Advanced Pain Assessment Using ECG Signals via Continuous Wavelet Transform Images

Sensors, Journal Year: 2025, Volume and Issue: 25(5), P. 1436 - 1436

Published: Feb. 26, 2025

Our research introduces a framework that integrates edge computing, quantum transfer learning, and federated learning to revolutionize pain level assessment through ECG signal analysis. The primary focus lies in developing robust, privacy-preserving system accurately classifies levels (low, medium, high) by leveraging the intricate relationship between perception autonomic nervous responses captured signals. At heart of our methodology processing approach transforms one-dimensional signals into rich, two-dimensional Continuous Wavelet Transform (CWT) images. These transformations capture both temporal frequency characteristics pain-induced cardiac variations, providing comprehensive representation different intensities. processes these CWT images sophisticated quantum–classical hybrid architecture, where devices perform initial preprocessing feature extraction while maintaining data privacy. cornerstone is Quantum Convolutional Hybrid Neural Network (QCHNN) harnesses entanglement properties enhance detection classification robustness. This quantum-enhanced seamlessly integrated framework, allowing distributed training across multiple healthcare facilities preserving patient privacy secure aggregation protocols. QCHNN demonstrated remarkable performance, achieving accuracy 94.8% assessment, significantly outperforming traditional machine approaches.

Language: Английский

---

An Energy-Efficient Decentralized Federated Learning Framework for Mobile-IoT Networks

Computer Networks, Journal Year: 2025, Volume and Issue: unknown, P. 111233 - 111233

Published: March 1, 2025

Language: Английский

---

Distributed Photovoltaic Short-Term Power Prediction Based on Personalized Federated Multi-Task Learning

Energies, Journal Year: 2025, Volume and Issue: 18(7), P. 1796 - 1796

Published: April 3, 2025

In a distributed photovoltaic system, data are affected by heterogeneity, which leads to the problems of low adaptability and poor accuracy power prediction models. This paper proposes scheme based on Personalized Federated Multi-Task Learning (PFL). The federal learning framework is used enhance privacy improve model’s performance in environment. A multi-task module added PFL solve problem that an FL single global model cannot all stations. cbam-itcn algorithm was designed. By improving parallel pooling structure time series convolution network (TCN), improved (iTCN) established, channel attention mechanism CBAMANet highlight key meteorological characteristics’ information feature extraction ability prediction. experimental analysis shows CBAM-iTCN 45.06% 42.16% lower than traditional LSTM, Mae, RMSE. Compared with FL, MAPE proposed this reduced 9.79%, for plants large deviation, experiences 18.07% reduction.

Language: Английский

---

Enhanced healthcare using generative AI for disabled people in Saudi Arabia

Alexandria Engineering Journal, Journal Year: 2025, Volume and Issue: 124, P. 265 - 272

Published: April 4, 2025

Language: Английский

---