Blockchain-Based Two-Stage Federated Learning With Non-IID Data in IoMT System

IEEE Transactions on Computational Social Systems, Journal Year: 2022, Volume and Issue: 10(4), P. 1701 - 1710

Published: Nov. 21, 2022

The Internet of Medical Things (IoMT) has a bright future with the development smart mobile devices. Information technology is also leading changes in healthcare industry. IoMT devices can detect patient signs and provide treatment guidance even instant diagnoses through technologies, such as artificial intelligence (AI) wireless communication. However, conventional centralized machine learning approaches are often difficult to apply within because difficulty large-scale collection data potential risk privacy breaches. Therefore, we propose blockchain-based two-stage federated approach that allows train global model collaboratively without gathering central server. Specifically, address problem poor training performance on non-independent identically distributed (non-IID) data, design data-sharing scheme significantly improve model's accuracy threatening user privacy. We client selection mechanism further system's efficiency. Finally, validate feasibility effectiveness our system simulation experiments three popular datasets (i.e., MNIST, Fashion-MNIST, CIFAR-10).

Language: Английский

---

Blockchain Meets Federated Learning in Healthcare: A Systematic Review With Challenges and Opportunities

IEEE Internet of Things Journal, Journal Year: 2023, Volume and Issue: 10(16), P. 14418 - 14437

Published: March 31, 2023

Recently, innovations in the Internet of Medical Things (IoMT), information and communication technologies, machine learning (ML) have enabled smart healthcare. Pooling medical data into a centralized storage system to train robust ML model, on other hand, poses privacy, ownership, regulatory challenges. Federated (FL) overcomes prior problems with aggregator server shared global model. However, there are two technical challenges: 1) FL members need be motivated contribute their time effort 2) may not accurately aggregate Therefore, combining blockchain can overcome these issues provide high-level security privacy for healthcare decentralized fashion. This study integrates emerging FL, We describe how blockchain-based plays fundamental role improving competent healthcare, where edge nodes manage avoid single point failure, while IoMT devices employ use dispersed clinical fully. discuss benefits limitations both technologies based content analysis approach. emphasize three main research streams systematic blockchain-empowered: IoMT; electronic health records (EHRs) (EMRs) management; 3) digital systems (internal consortium/secure alerting). In addition, we present novel conceptual framework blockchain-enabled environment. Finally, highlight challenges future directions applications.

Language: Английский

---

DMFL_Net: A Federated Learning-Based Framework for the Classification of COVID-19 from Multiple Chest Diseases Using X-rays

Sensors, Journal Year: 2023, Volume and Issue: 23(2), P. 743 - 743

Published: Jan. 9, 2023

Coronavirus Disease 2019 (COVID-19) is still a threat to global health and safety, it anticipated that deep learning (DL) will be the most effective way of detecting COVID-19 other chest diseases such as lung cancer (LC), tuberculosis (TB), pneumothorax (PneuTh), pneumonia (Pneu). However, data sharing across hospitals hampered by patients' right privacy, leading unexpected results from neural network (DNN) models. Federated (FL) game-changing concept since allows clients train models together without their source with anybody else. Few studies, however, focus on improving model's accuracy stability, whereas existing FL-based detection techniques aim maximize secondary objectives latency, energy usage, privacy. In this work, we design novel model named decision-making-based federated (DMFL_Net) for medical diagnostic image analysis distinguish four distinct disorders including LC, TB, PneuTh, Pneu. The DMFL_Net has been suggested gathers variety hospitals, constructs using DenseNet-169, produces accurate predictions information kept secure only released authorized individuals. Extensive experiments were carried out X-rays (CXR), performance proposed was compared two transfer (TL) models, i.e., VGG-19 VGG-16 in terms (ACC), precision (PRE), recall (REC), specificity (SPF), F1-measure. Additionally, also default FL configurations. + DenseNet-169 achieves an 98.45% outperforms approaches classifying successfully protects privacy among diverse clients.

Language: Английский

---

Enhancing Internet of Medical Things security with artificial intelligence: A comprehensive review

Computers in Biology and Medicine, Journal Year: 2024, Volume and Issue: 170, P. 108036 - 108036

Published: Jan. 28, 2024

Over the past five years, interest in literature regarding security of Internet Medical Things (IoMT) has increased. Due to enhanced interconnectedness IoMT devices, their susceptibility cyber-attacks proportionally escalated. Motivated by promising potential AI-related technologies improve certain cybersecurity measures, we present a comprehensive review this emerging field. In review, attempt bridge corresponding gap modern that deploy AI techniques performance and compensate for privacy vulnerabilities. direction, have systematically gathered classified extensive research on topic. Our findings highlight fact integration machine learning (ML) deep (DL) improves both measures speed, reliability, effectiveness. This may be proven useful improving devices. Furthermore, considering numerous advantages as opposed core counterparts, including blockchain, anomaly detection, homomorphic encryption, differential privacy, federated learning, so on, provide structured overview current scientific trends. We conclude with considerations future research, emphasizing AI-driven landscape, especially patient data protection data-driven healthcare.

Language: Английский

---

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

---

Blockchain-Based Federated Learning With Secure Aggregation in Trusted Execution Environment for Internet-of-Things

IEEE Transactions on Industrial Informatics, Journal Year: 2022, Volume and Issue: 19(2), P. 1703 - 1714

Published: April 26, 2022

This paper proposes a blockchain-based Federated Learning (FL) framework with Intel Software Guard Extension (SGX)-based Trusted Execution Environment (TEE) to securely aggregate local models in Industrial Internet-of-Things (IIoTs). In FL, can be tampered by attackers. Hence, global model generated from the erroneous. Therefore, proposed leverages blockchain network for secure aggregation. Each node hosts an SGX-enabled processor that performs FL-based aggregation tasks generate model. Blockchain nodes verify authenticity of aggregated model, run consensus mechanism ensure integrity and add it distributed ledger tamper-proof storage. cluster obtain its before using it. We conducted several experiments different CNN datasets evaluate performance framework.

Language: Английский

---

On the Physical Layer Security of Federated Learning Based IoMT Networks

IEEE Journal of Biomedical and Health Informatics, Journal Year: 2022, Volume and Issue: 27(2), P. 691 - 697

Published: May 10, 2022

Internet of Medical Things (IoMT) connects different medical devices, health sensors and hospital records to data platforms using wireless communications. Federated Learning (FL) is an emerging collaborative learning technique that can be beneficial for IoMT due reduced communication overhead enhanced security. This paper provides overview architectures used in FL potential approaches based IoMT. We also discuss how Physical Layer Security (PLS) efficient privacy preservation highlight the recent work this area major research challenges related PLS assisted provide a case study demonstrating clustering devices (such single device each cluster acts as head) enhances secrecy rate network compared its non-clustered counterpart. Finally, we future opportunities open questions

Language: Английский

---

Secure IoMT for Disease Prediction Empowered With Transfer Learning in Healthcare 5.0, the Concept and Case Study

IEEE Access, Journal Year: 2023, Volume and Issue: 11, P. 39418 - 39430

Published: Jan. 1, 2023

Identifying human diseases remains a difficult process, even in the age of advanced information technology and smart healthcare industry 5.0. In 5.0, precise prediction diseases, particularly lethal cancer is critical for well-being. The global Internet Medical Things sector has at breakneck pace recent years, from small wristwatches to large aircraft. aspects include security privacy, owing massive scale deployment networks. Transfer learning with secure IoMT-based approach considered. google net deep machine-learning model used accurate disease We can easily reliably anticipate body by using transfer approach. Furthermore, results proposed techniques are validate best methodology reached 98.8%, better than state-of-the-art methodologies previously

Language: Английский

---

A systematic review of privacy-preserving methods deployed with blockchain and federated learning for the telemedicine

Healthcare Analytics, Journal Year: 2023, Volume and Issue: 3, P. 100192 - 100192

Published: May 5, 2023

The unexpected and rapid spread of the COVID-19 pandemic has amplified acceptance remote healthcare systems such as telemedicine. Telemedicine effectively provides communication, better treatment recommendation, personalized on demand. It emerged possible future medicine. From a privacy perspective, secure storage, preservation, controlled access to health data with consent are main challenges effective deployment is paramount fully overcome these integrate telemedicine system into healthcare. In this regard, emerging technologies blockchain federated learning have enormous potential strengthen system. These help enhance overall standard when applied in an integrated way. primary aim study perform systematic literature review previous research privacy-preserving methods deployed for This in-depth qualitative analysis relevant studies based architecture, mechanisms, machine used access, analytics. survey allows integration suitable techniques design secure, trustworthy, accurate model guarantee.

Language: Английский

---

Fed-Inforce-Fusion: A federated reinforcement-based fusion model for security and privacy protection of IoMT networks against cyber-attacks

Information Fusion, Journal Year: 2023, Volume and Issue: 101, P. 102002 - 102002

Published: Sept. 2, 2023

Language: Английский

---