Integrating open-source medical systems, with advancements in 3D printing technology and microcomputer systems such as Arduino and Raspberry Pi, has revolutionized the healthcare industry. However, it has also exposed cybersecurity vulnerabilities in hospitals. This paper presents a web-based botnet as a proof-of-concept to demonstrate potential disruptions in the control flow of a syringe pump in an IoT medical network testbed. Our lightweight botnet stands out for its rapid deployment and minimal use of resources. We also provide a publicly available dataset from this botnet for cybersecurity research on open-source medical systems. Additionally, we developed a methodology for feature selection to detect botnet attacks. Our comparative study with various machine learning algorithms revealed the best strategy for detecting these attacks using network traffic data from benign and malicious environments. The results were impressive, with our feature selection technique achieving over 99% accuracy on the testing dataset, successfully identifying 63,380 out of 63,382 attack instances.
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号