Cyber-Security Threats to IoMT-Enabled Healthcare Systems
S. Roobini, M. Kavitha, M. Sujaritha, Rajesh Kumar Dhanaraj · 2022
The traditional healthcare systems are confronted with new issues as the number of patients keeps rising. The design of the Internet of Medical Things (IoMT) addresses this problem along with improving the healthcare system’s accuracy, dependability, productivity, and efficacy. IoMT devices are subjected to network interruption and denial of service assaults as a result of these attacks. IoMT may be viewed as an improvement and expenditure in order to respond to patient requirements more effectively and efficiently. An enormous number of medical devices are introduced into the healthcare system, which makes it a challenging task to identify traditional methods of attack. Deep learning is an approach to design models of optimal security based on actual device data. Using the N-BaIoT dataset, botnet attacks are our main focus against various IoMT devices that come in a variety of shapes and sizes and deep learning models that are based on it are constructed, for each gadget type. The findings of the experiment demonstrate that incorporating features decreases training time and model complexity while increasing bot detection rates. Our model delivers greater precision and accuracy when compared to other neural network approaches. There is a need to handle security in order to assure user confidence, encourage wide acceptance, and realize the benefits of IoMT. These security issues highlight an increasing trend of cyber-assaults on system infrastructure, as well as system inherent weaknesses inside the environment.