The Internet of Medical Things (IoMT) has transformed healthcare by allowing real-time patient monitoring, remote diagnoses, and effective data interchange. The increasing reliance on interconnected medical equipment has increased cybersecurity risks for healthcare organizations. This survey offers an extensive examination of Intrusion Detection Systems (IDSs) targeted for IoMT contexts. This survey emphasizes the proposed methods that used to build IDS, classifying them into machine learning (ML), deep learning (DL), fuzzy logic (FL), and hybrid approaches for safeguarding healthcare networks. This paper investigates the IoMT architecture, identifies security concerns across multiple tiers, and analyzes potential vulnerabilities including denial-of-service attacks, ransomware, and man-in-the-middle attacks. The research highlights the significance of IDSs in alleviating cyber threats and protecting sensitive medical information through a comparison of cutting-edge methodologies. We outline significant issues that persist and emphasize domains requiring additional research to enhance the security and resilience of IoMT systems.

doi:10.5121/ijaia.2024.15604

Volume 15, Number 4

AI for IOMT Security: a Survey of Intrusion Detection Systems, Architectures, Attacks and Challenges

Authors

Ghaida Balhareth, Mohammad Ilyas , and Basmh Alkanjar, Florida Atlantic University, USA

Abstract

The Internet of Medical Things (IoMT) has transformed healthcare by allowing real-time patient monitoring, remote diagnoses, and effective data interchange. The increasing reliance on interconnected medical equipment has increased cybersecurity risks for healthcare organizations. This survey offers an extensive examination of Intrusion Detection Systems (IDSs) targeted for IoMT contexts. This survey emphasizes the proposed methods that used to build IDS, classifying them into machine learning (ML), deep learning (DL), fuzzy logic (FL), and hybrid approaches for safeguarding healthcare networks. This paper investigates the IoMT architecture, identifies security concerns across multiple tiers, and analyzes potential vulnerabilities including denial-of-service attacks, ransomware, and man-in-the-middle attacks. The research highlights the significance of IDSs in alleviating cyber threats and protecting sensitive medical information through a comparison of cutting-edge methodologies. We outline significant issues that persist and emphasize domains requiring additional research to enhance the security and resilience of IoMT systems.

Keywords

Machine learning, Deep learning, Fuzzy logic, Intrusion Detection Systems (IDS), Internet of Medical Things (IoMT).