Towards Federated Learning Intrusion Detection System (IDS) within Internet of Medical Things (IoMT) Ecosystems
Open AccessThis research demonstrates how federated learning (FL), and architecture could be leveraged across Internet of Medical Things (IoMT) ecosystems with Intrusion detection systems detect and mitigate cyber threats. The IoMT is comprised of heterogenous devices such as wearable, implants and on-person sensors that transmit and receive medical data. These devices are vulnerable to diverse types of security threats, and thus present a significant risk to a patient’s privacy and safety. Because security is a critical factor for successfully merging IoMT into pervasive healthcare systems, there is an urgent need for new security mechanisms to prevent threats on the IoMT edge network. Federated learning (FL) provides enhanced privacy as compared to traditional centralized learning models. This is extremely important in healthcare ecosystems to provide protection on Personally Identifiable Information (PII) and Personal Health Information (PHI) data which cannot be delivered with traditional centralized methods. Additionally, implementing decentralized solutions provides enhanced optimization techniques not prevalent in centralized learning traditional methods. (R. Durga, 2021) The integration of blockchain provides a fully decentralized ecosystem aimed to provide resiliency and immutable protections on personal data. The research analyzes cyber threat detection using centralized and federated learning models. Fog nodes are employed with IoMT gateways to minimize data in transit and accelerate detection of cyber threats which may impact local IoMT devices. Hyperledger Fabric Blockchain is used to mitigate the FL trust and security shortcomings. (Sun J, 2022) Due to the limited publicly available healthcare datasets, this research will explore analysis on WUSTL-EHMS-2020 dataset to provide an enriched solution comprised of biometric and network attack data. The rationale for using this dataset is that IoMT is a subset of the broader IoT ecosystem. (Manickam P, 2022) Several threat vectors associated with IoT are prevalent within healthcare IoMT. Our focus is to identify cyber threats within the healthcare network that provide services consumed by providers for patient diagnosis. Leveraging machine learning techniques, the research aims to provide models that are equivalent or better efficacy as compared to centralized techniques. Enhancing privacy is an important characteristic of the healthcare domain. The results will aim to demonstrate the feasibility of using Federated Learning (FL) as compared to CL methods supported by blockchain to enhance privacy IoMT based systems which have reduced computational and network resources than a traditional centralized approach.
- All rights reserved
Notice to Authors
If you are the author of this work and you have any questions about the information on this page, please use the Contact form to get in touch with us.