Jul 2026· International Conference Computing Methodologies and Communication· pp. 734-738· 0 citations· 17 references
Abstract
Using federated learning (FL) in health care applications, the teams will be able to collaborate to predict the health of patients without exchanging any personally identifiable information. Regardless of this benefit, FL is still not widely used since it is susceptible to serious privacy and security risks like membership inference and model poisoning attacks. To address these concerns, this paper proposes a blockchain-enhanced FL model that involves safe aggregation and audit-trail immutability to thwart privacy leakage and manipulation of medical data analysis by opponents. Random Forest and LightGBM classifiers are trained on fake healthcare datasets in various scenarios of attack. Some of the security measures that are used to assess the system include attack success rate, cost of communication, differential privacy (DP) epsilon values and blockchain-based tamper detection delay. The proposed approach can achieve measurable privacy leaks reduction and a 40-percent lower success rate in membership inference attacks, as demonstrated in experiments. Also, blockchain audit trails can provide real-time resilience against tampering of data. According to these findings, blockchain-enhanced FL appears to be a promising framework of medical AI systems capable of ensuring the patient data safety and privacy.
The proposed framework effectively integrates encryption, federated intrusion detection, explainable artificial intelligence, and blockchain security to enhance privacy, transparency, and reliability in IoMT healthcare networks.
P. Banupriya, K. Vanitha· Journal of Vibration Enginee...· 0 citations
The privacy of sensitive medical information is essential since the application of effective analytics is crucial to identifying healthcare fraud. Regulatory and security limitations do not always allow the use of the traditional centralized approaches. This survey presents the current trend in privacy-conscious fraud detection, and it pays a lot of attention to the application of trust-centric federated learning along with blockchain and explainable artificial intelligence. It talks about the potential of the decentralized learning systems to train models without sharing raw data, and blockchain can be employed to ensure transparency and auditability. Also, the issue of explainable models in increasing interpretability and regulatory compliance is addressed. The essential issues are examined, i.e., adversarial robustness, communication overhead, and system scalability. The questionnaire concludes with the future research directions of developing safe, trusted, and effective healthcare fraud detection systems.
Jeevarathinam M, Kumaravelan G, Lakshmi Narayanan H· 2026 4th International Confe...· 0 citations
A block chain-based Privacy-preserving and Secure Federated Learning (BPS-FL) system that uses threshold homomorphic encryption to safeguard the local gradients of clients in order to successfully solve such privacy and security assault challenges is suggested.
Umema Samreen, I. S. P. James· American Journal of AI Cyber...· 0 citations
The proposed framework for financial system fraud detection that is safe and protects privacy while resolving issues with data sharing, legal restrictions, and cybersecurity threats is appropriate for practical financial applications since it successfully improves fraud detection while guaranteeing Privacy Preservation, security, and openness.
The fast development of Internet of Medical Things (IoMT) devices has also compounded the necessity to have secure, privacy-sensitive, and efficient Electronic Health Record (EHR) sharing mechanisms. To overcome the fundamental issues associated with the IoMT ecosystems of data privacy leakage, excessive communication overhead, and inefficient decentralized learning, this paper suggests a Privacy-Aware Blockchain and Federated Learning (PABFL) system. The suggested framework combines a blockchain-based secure data sharing with federated learning-based distributed model training, which guarantees data confidentiality and integrity without the centralized storage. It is tested with NVIDIA Jetson AGX Xavier nodes, which have CUDA 11.4 acceleration and is tested on both CardioFit and GlucoWatch datasets. Through experiments, it is shown that PABFL greatly outperforms more traditional methods like Federated Averaging (FedAvg), Federated Proximal (FedProx), and Asynchronous Advantage Actor-Critic (A3C) with an accuracy difference of 6.8%-9.5%, energy consumption of 18.2%-24.7%, training time of 21.4%-29.3% These improvements are explained by optimized model aggregation, safe lightweight blockchain operations, and resource-efficient edge device utilization. The suggested framework provides high-quality, scalable, and privacy-friendly EHR sharing that is very appropriate in real-time healthcare applications in IoMT settings and helps to develop safe and smart digital healthcare systems.
G. Prasad, M. V. Rao· International Conference Com...· 0 citations
The proposed hybrid approach outperforms ML-only and blockchain-only baselines, offering a scalable, secure, and real-time IDS for IIoT infrastructures.
