Enhancing IoMT Security: A Conception of RFE-Ridge and ML/DL for Anomaly Intrusion Detection

The outbreaks of smart cities have delivered smart connectivity of numerous Internet of Things (IoT) resulting in the boom of the Internet of Medical Things (IoMT), which delivers enhanced treatments and improves patient healthcare. However, the 2017 “WannaCry” ransomware cyber-attack in the United Kingdom compromised privacy and suspended operations at 48 healthcare providers. Despite the mammoth demand for IoMT devices, cyber-assaults on connected healthcare systems can threaten patient’s lives and can also tamper healthcare data. This paper proposes a conception of RFE-Ridge feature selection empowered with Machine/Deep learning models for implementing a viable intrusion detection in the IoMT system, meanwhile, a comparison between ML/DL models was conducted in terms of advantages and limitations. It is noteworthy that the proposed framework can be employed to construct an effective Intrusion Detection System (IDS) that strengthens the security of the IoMT against pervasive cyber-attacks.