Safeguarding IoMT: Semi-automated Intrusion Detection System (SAIDS) for detecting multilayer attacks

The Internet of Medical Things (IoMT) plays a significant role in the healthcare system as it improves effectiveness and efficiency of treatment by continuously monitoring patients using smart home sensor and wearables (Fig. 1).

IoMT devices are vulnerable to Multi-layer attacks that are
exploiting multiple layers of IoMT architecture (Fig. 2). Denial-of-service (DoS) and Man-In-The-Middle (MITM) attacks, for instance, can target the three layers of the IoMT system and lead to serious consequences, such as theft of patients’ sensitive data and reputational damages [2].

This project aims to create a robust detection system for multilayer attacks using a Semi-automated Intrusion Detection System (SAIDS) for IoT devices.
To achieve this aim, we have focused on the following objectives:
• Explore a variety of feature selection algorithms.
• Apply feature weighting.
• Integrating human and machine learning approaches to work together.
• Increase detection efficiency by utilizing significant features.

[1] Al Sukhni, B., Dave, J.M., Manna, S.K. and Zhang, L. (2022) Investigating the security issues of multi-layer IoT attacks using machine learning techniques. IEEE, pp. 1.
[2] Al Sukhni, B., Manna, S.K., Dave, J.M. and Zhang, L. (2022) Machine learning-based solutions for securing IoT systems against multilayer attacks. Springer, pp. 140.