A Blockchain-Federated Learning for Privacy-Preserving Intrusion Detection in IoMT

Abstract

This project aims to develop an intelligent intrusion detection system for Internet of Medical Things (IoMT) environments by integrating three key technologies: Artificial Intelligence, Federated Learning, and Blockchain. Initially, centralized learning models were adopted; however, they demonstrated limitations in preserving data privacy and posed a single point of failure that threatens system stability. To address these challenges, Federated Learning was employed as an alternative that enables training models locally on edge devices without sharing raw data, thereby enhancing privacy and reducing reliance on centralized servers. Nevertheless, the presence of a central server in traditional federated learning remains a critical security vulnerability. Therefore, Blockchain technology was integrated to provide a decentralized and secure infrastructure for transparently exchanging and verifying model updates. This integration has contributed to enhancing the security and reliability of the system, mitigating the risks of cyberattacks, and offering a promising solution for securing intelligent healthcare systems based on IoMT technologies.