Intelligent AI-based Healthcare Cyber Security System using Multi-Source Transfer Learning Method

Abstract

Cyber-security intelligence have made a great impact over healthcare industry where several researchers are developing new techniques to improve security for healthcare systems. Besides, Artificial Intelligence (AI) become the tremendous technology in recent decades to improve the existing methods to be more intelligent. In this paper, we proposed cyber attack detection system for healthcare sector with centralized and federated transfer learning mode. Edge of Things (EoT) framework is developed in connection with cloud and healthcare sectors to transmit the data efficiently and the proposed Centralized with Multi-Source Transfer Learning (CMTL) algorithm which is used for detection and classification of various threats such as information gathering, DoS/DDoS attacks, Malware attacks, Injection attacks, and Man in the Middle attacks. Performance of the proposed framework is evaluated using various datasets such as EMNIST, X-IIoTID, and Federated TON_IoT. Our framework outperforms with the analysis of execution time and obtains high level accuracy when compared with different algorithms.