Digital Twin Enhanced Data Protection Based on Cloud-Edge Collaboration in Healthcare System

Empowered with sensing networks and edge computing, the Healthcare system represented by the health monitoring system (HMS) gradually becomes an important part of future medical technology. HMS has been gradually applied to practice scenarios, occurring that the resource-restricted Internet of Things (IoT) nodes and heterogeneous networks' structure make data protection more difficult. Since anomalies of HMS commonly cause extensive loss even human safety, timely anomaly prediction schemes are regarded as valuable to ensure the data protection of HMS. Most of the existing solutions are based on historical logs, which makes it difficult to identify new faults and easily causes extensive security threats. To solve this problem, we proposed a distributed digital twins (DT) enhanced abnormal prediction scheme based on cloud-edge collaboration to improve the availability and data protection of the HMS. This work proposed an HMS framework based on distributed DT to take full advantage of the parallel simulating capabilities of DT. In addition, considering the limitation of user demand and marginal server resources, we further proposed customized DT simulation operating mechanisms for medical devices in the HMS environment. The feasibility and efficiency of this work were discussed and proved in the analysis and performance evaluation.