A Deep Learning-Based Framework to Protect Healthcare Systems from Cyber-Attacks

Abstract

The Internet of Medical Things (IoMT) enhances healthcare systems by improving scalability, reliability, effectiveness, and accuracy through interconnected medical devices that enable remote monitoring and realtime patient data analysis. However, IoMT networks are vulnerable to security threats like man in the middle attacks, replay attacks, remote hijacking, malware, and Denial-of-Service (DoS) attacks due to resource constraints and heterogeneity. To address these challenges, this paper proposes a deep learning-based Intrusion Detection System (IDS) that secures healthcare systems by incorporating two key components: an attack detection framework for IoMT devices and a malware detection method for Windows Portable Executable (PE) files used by medical staff computers and data servers. The proposed model utilizes a Recurrent Neural Network (RNN) to accurately detect and identify threats with low computational cost, achieving 99.35% accuracy on the "Portable Executable (PE) malware" dataset and 99.80% accuracy on the "IoT-Healthcare security" dataset, demonstrating superior performance over existing methods.