Machine Learning-Based Detection for Cyber Attacks in Internet of Medical Things Devices

The Internet of Medical Things (IoMT) represents a technological advancement in healthcare, facilitating real-time patient monitoring and data-driven decision-making. However, the increased connectivity of medical devices introduces cybersecurity risks, particularly from DoS and ARP Spoofing attacks. In this study we investigate the application of various machine learning (ML) models to detect these cyber threats on Bluetooth, WiFi, and Message Queuing Telemetry Transport (MQTT)enabled devices. By leveraging the CIC IoMT 2024 dataset, we evaluate the performance of Decision Trees, Random Forests, Gradient Boosting, XGBoost, Recurrent Neural Networks, and Isolation Forests. The findings indicate that while Decision Trees and Random Forests exhibit high accuracy in detecting DoS and ARP Spoofing attacks, the complexity and volume of features influence model performance. This research highlights the necessity of robust cybersecurity measures to ensure the security of medical services.