Explainable Learning Machines for Securing the IoMT Networks

Abstract

The necessity to improve modern medical networks such as internet of medical things (IoMT) to monitor patients and their health condition has raised due to the effects of population ageing, increasing number of patients, deficiency of treatment facilities, and spread of widespread diseases. However, resisting cyber-attacks is a challenging concern for researchers. This chapter proposes an explainable learning machines-based security framework for detecting cyber-attacks against IoMT networks in real-time. The proposed model is based on the phenomenon of extreme learning machines to detect multiple kinds of cyber-space attacks carried against IoMT systems. The authors also explain the detection decisions to expand trust management in the employed machine learning algorithm and facilitate security professionals to comprehend the undiscovered data evidence and causal inference. Experiments show the effectiveness of the proposed approach signifying its utility as a workable security framework in contemporary networks of IoMT-based healthcare systems.