Transformer-Based Intrusion Detection for Securing Medical Applications in 5G IoMT Networks

Protecting medical network from cyber-attacks is important as the Internet of Medical Things expands into many verticals and increases use of connected industrial devices in the future. The sheer volume of data created by medical environment devices has not only pushed the necessity for Artificial Intelligence as a secondary to traditional cybersecurity, but as a supplement to traditional cybersecurity. Our key contribution is to introduce real time analysis of industrial network traffic, enabling the exploration of feasibility of cyberattacks on industrial networks, by employing the power of advanced sequence modeling capabilities in transformer architectures like state of the art, and as such provide intriguing insights into explainability.