Robust and Reliable Security Approach for IoMT: Detection of DoS and Delay Attacks through a High-Accuracy Machine Learning Model

Abstract

Internet of Medical Things (IoMT ) refers to the network of medical devices and healthcare systems that are connected to the internet. However, this connectivity also makes IoMT vulnerable to cyberattacks such as DoS and Delay attacks , posing risks to patient safety, data security, and public trust. Early detection of these attacks is crucial to prevent harm to patients and system malfunctions. In this paper, we address the detection and mitigation of DoS and Delay attacks in the IoMT using machine learning techniques. To achieve this objective, we constructed an IoMT network scenario using Omnet++ and recorded network traffic data. Subsequently, we utilized this data to train a set of common machine learning algorithms. Additionally, we proposed an Enhanced Random Forest Classifier for Achieving the Best Execution Time (ERF-ABE), which aims to achieve high accuracy and sensitivity as well as low execution time for detecting these types of attacks in IoMT networks. This classifier combines the strengths of random forests with optimization techniques to enhance performance. Based on the results, the execution time has been reduced by implementing ERF-ABE, while maintaining high levels of accuracy and sensitivity.