Secure E-healthcare System Based on Blockchain Technology and Ensemble Models

Abstract

Ensuring the security and privacy of e-healthcare systems is now crucial due to the increasing occurrence of various cyberattacks in vulnerable environments, including Distributed Denial of Service (DDoS). Blockchain technology and ensemble models offer one of the most promising solutions to ensure data privacy and integrity within secure e-healthcare system. In this paper, blockchain and ensemble models' capabilities are employed to develop a secure e-healthcare system. The DDoS attack detection mechanism utilizes ensemble model training and smart contracts to significantly enhance attacks detection efficiency and accuracy. We assess the approach's performance by evaluating two ensemble models, Random Forest (RF) and Extreme Gradient Boosting (XGBoost) focusing on DDoS attacks detection using the recent dataset CICIoMT2024. The combination of blockchain and ensemble models result in accurate detection of DDoS attacks exceeding 90%. This improvement signifies an unprecedented innovation in ensemble models in blockchain-enabled e-healthcare systems. Moreover, in terms of prediction time, the DT model excels with the fastest prediction time.