ECU-IoHT: A dataset for analyzing cyberattacks in Internet of Health Things

In recent times, cyberattacks on the Internet of Health Things (IoHT) have continuously been growing, and so it is important to develop robust countermeasures. However, there is a lack of publicly available datasets reflecting cyberattacks on IoHT, mainly due to privacy concerns. This paper showcases the development of a dataset, ECU-IoHT, which builds upon an IoHT environment having different attacks performed that exploit various vulnerabilities. This dataset was designed to help the healthcare security community in analyzing attack behavior and developing robust countermeasures. No other publicly available datasets have been identified for cybersecurity in this domain. Anomaly detection was performed using the most common algorithms, and showed that nearest neighbor-based algorithms can identify attacks better than clustering, statistical, and kernel-based anomaly detection algorithms.

Automated summary

This paper presents the development of a new dataset, ECU-IoHT, to assist the healthcare security community in analyzing attack behavior and developing robust countermeasures. The study found that nearest neighbor-based algorithms outperformed clustering, statistical, and kernel-based anomaly detection algorithms in identifying cyberattacks.