Advancements in Machine Learning-Based Intrusion Detection in Iomt: Research Trends and Challenges

Abstract

The integration of smart medical technologies and connected applications into the Internet of Medical Things (IoMT) is revolutionizing healthcare by optimizing patient connectivity, immediate monitoring, and data mining. However, the openness of IoMT networks exposes these systems to a variety of cyber threats. These attacks jeopardize the protection, availability, and accuracy of sensitive information and associated resources. It is therefore essential to strengthen the security of IoMT infrastructures in the face of these malicious activities. Identifying new attacks in these environments is a major challenge for researchers and industry professionals. This paper analyzes machine learning-based strategies for identifying IoMT intrusions. We conduct a comprehensive examination of existing solutions, focusing on the analytical methodologies employed, the datasets utilized, and the effectiveness of detection performance. In addition, we discuss the challenges and considerations involved in designing high-performance systems capable of detecting cyberattacks targeting the IoMT. Finally, we highlight future prospects for the development of advanced protection mechanisms tailored to connected medical infrastructures.