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

A Survey on Heterogeneity Taxonomy, Security and Privacy Preservation in the Integration of IoT, Wireless Sensor Networks and Federated Learning

“…In this section, we will explore the basic principles of FL, including model aggregation, privacy preservation, and communication protocols. Additionally, we will discuss the challenges and opportunities of implementing FL in IoT systems [19,51,67]. The general FL process includes the following key steps: system initialisation and device selection, where the aggregator chooses an IoT task such as human activity recognition and sets up learning parameters, e.g., learning rates and communication rounds; distributed local training and updates, where after the training configuration, the server initialises a new model and transmits it to the IoT clients to start the distributed training; finally, model aggregation and download, where after collecting all model updates from local clients, the server aggregates them and calculates a new version of the global model.…”

“…Privacy Preservation: FL enables the training of models on sensitive data without compromising user privacy. This is particularly important in healthcare applications where personal health data needs to be protected [67,68]. To be specific, FL clients do not need to upload their data to the central server, and thus the vulnerabilities to security threats can be minimised, since the model updates are ephemeral and anonymous.…”

Cybersecurity for Sustainable Smart Healthcare: State of the Art, Taxonomy, Mechanisms, and Essential Roles

“…The emerging technologies in SHSs collect patients' sensitive healthcare data and vital parameters, which are stored in cloud databases for healthcare professionals to share and analyze. However, their security and privacy remain serious concerns [31]. Privacy in healthcare refers to protecting patients' healthcare data from unauthorized access, use, and disclosure to third parties.…”

“…Some of the common cyber threats and attacks in smart healthcare include healthcare data breaches, privacy concerns, denial-of-service (DoS) and distributed DoS (DDoS) attacks, ransomware, phishing attacks, eavesdropping attacks, man-in-the-middle attacks, impersonation attacks, insider threats, replay attacks, medical identity thefts, brute-force attacks, fake base stations, supply chain attacks, medjacking, advanced persistent threats, SQL injection attacks, legacy systems, side-channel attacks, jamming attacks, buffer overflow, Sybil attacks, routing attacks, cross-site scripting attacks, cross-site request forgery attacks, session hijacking attacks, account hijacking, cookie manipulation attacks, sensor attacks, tampering attacks, zeroday vulnerabilities, cryptographic attacks, stolen physical smart device attacks, cloud-based threats, medical IoT device vulnerabilities, attacks associated with blockchain, evasion attacks, poisoning attacks, extraction attacks/model stealing/model inversion, and regulatory compliance challenges [24][25][26][27][28][29]. These attacks target patients' health information, financial information (e.g., credit card and bank account numbers), patients' identifying information (e.g., social security numbers), and medical research and innovation intellectual property, thus compromising privacy, confidentiality, access control, integrity, authentication, nonrepudiation, anonymity, and availability [30][31][32]. Between March 2022 and March 2023, data breaches in the healthcare industry cost nearly US$11 million [33].…”

An Efficient Multi-User Groupwise Integrity CP-ABE(GI-CPABE) for Homogeneous and Heterogeneous Cloud Blockchain Transactions

“…Patients who record their health information on a blockchain may have the option to grant access to their electronic health data to other parties, preventing unauthorized tampering. In essence, blockchain serves as a distributed ledger for sharing and trading health information among involved parties [21]. In this context, authorized organizations can access the blockchain for inspection and block validation, in contrast to public blockchains like Bitcoin.…”