A Federated Graph-Based Intrusion Detection Framework for Unknown Attack Detection in Internet of Medical Things(IoMT)
Abstract
Abstract The integration of Internet of Things (IoT) into smart healthcare, known as the Internet of Medical Things (IoMT), has significantly improved patient monitoring and clinical decision-making. While the IoMT systems has ability to provide intelligent services,the interconnected nature of IoMT system has also make the system vulnerable to sophisticated cyberattacks and potential intrusions. Traditional deep learning based Intrusion Detection Systems (IDSs) serve as potential tools for cybersecurity infrastructure, safeguarding IoMT systems against evolving cyberthreats. However, existing deep learning architectures, particularly Convolutional Neural Networks (CNNs) approaches may suffer from the issue of limited capability of unseen attack detection and privacy preservation. To address this limitation, this paper presents a Federated Graph-Based Intrusion Detection framework that combines Graph Convolutional Networks (GCN), Graph Attention Networks (GAT), and Federated Learning(FL) to achieve privacy-preserving intrusion detection in IoMT environments. Unlike previous studies, the proposed work explicitly models network traffic as graphs, preserves patient privacy through federated learning, and evaluates robustness against unseen attacks through a realistic cross-dataset validation by training and validation on CICIoMT2024 and testing on independent Mirai attack samples from CIC-DIAD2024. The proposed model achieves 99.65% binary classification accuracy and 90.64%…
Read the paper
Related studies
- Active Learning during Federated Learning for Object Detection
- Fairness-aware federated learning
- Decentralized Federated Learning
- Building a Personally Identifiable Information Recognizer in a Privacy Preserved Manner Using Automated Annotation and Federated Learning
- Preservation of Patient Level Privacy: Federated Classification and Calibration Models
- Efficient explainable federated learning
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.