Abstract:
In smart healthcare systems, the security of medical sensor networks is critical due to the sensitive nature of patient data and the increasing threat of cyberattacks. However, current security measures often fail to fully protect these networks from unauthorized access and data breaches. To tackle these challenges, we propose an AI-powered security framework aimed at safeguarding patient data and enhancing trust in healthcare technology. Our system employs ensemble learning models, such as Gradient Boosting and Random Forest, to detect and mitigate cyber threats in medical sensor networks. These models, integrated with an Intrusion Detection System (IDS), actively monitor network traffic to identify any anomalous behaviour that could indicate potential security breaches. The system has been trained on the CICIoMT2024 dataset, which contains various attack scenarios targeting medical sensor networks. Our models have demonstrated up to 99% accuracy in detecting anomalies and ensuring effective threat mitigation. Furthermore, the system strengthens access control, blocking unauthorized access and preserving the integrity of patient data. Through thorough testing and validation, our AI-driven security framework has proven to enhance the security of smart healthcare systems, improve patient privacy, and foster greater
trust in healthcare technologies.
