An intelligent platform for threat assessment and cyber-attack mitigation in IoMT ecosystems

Nicholas Kolokotronis, Maria Dareioti, Stavros Shiaeles, Emanuele Bellini

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The increasing connectivity of medical devices along with the growing complexity, heterogeneity and attack surface of healthcare ecosystems has lead to numerous severe cyber-attacks. This paper proposes a novel collaborative security platform for threat assessment, intelligent detection and autonomous mitigation. The solution leverages machine learning(ML) and federated learning for detecting and preventing sophisticated multi-stage attacks, as well as blockchain for supporting integrity verification and accountability to defend against advanced persistent threats. The solution uses a distributed edge approach, performing intensive computations at the edge of the network, where information is generated, to achieve real-time processing of security events. The prevention capabilities employ autonomous decision-making with optimal response strategies towards cyber-attacks and runtime adaptation; these rely on dynamic risk-based models that use real-time information about security incidents.

Original languageEnglish
Title of host publication2022 IEEE Globecom Workshops (GC Wkshps)
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages541-546
ISBN (Electronic)9781665459754
ISBN (Print)9781665459761
DOIs
Publication statusPublished - 12 Jan 2023

Keywords

  • Graph neural network
  • human action recognition
  • motion prediction
  • skeleton model

Cite this