Intelligent digital twin reference architecture models for medical and healthcare industry

Document Type


Publication Title

Digital Twin for Healthcare: Design, Challenges, and Solutions


The recent redefinition of Digital Twin (DT) significantly has been expanding DT's potential and increasing momentum in both industries and academic research. DTs have not only widely adopted in manufacture industry but also become a new trend in medical and healthcare industry. However, traditional architectures do not provide satisfactory solutions for protecting privacy, breaking data silos, handling big data, and quickly integrating heterogeneous DTs. In this chapter, we introduced Intelligent DT architecture reference model for medical and health industry. It not only addresses the above challenges but also analytically defines the location of DT services deployed, the detail logic capabilities DT system provides and interfaces among them. We analytically molded DT's architecture models, analyzed their suitable application scenarios, and recommend using loosely coupled DT architecture model that combine data-centric, event-centric and application centric methodologies for complex cases in medical and healthcare industry. The architecture is helpful in identifying various levels of business and technical requirements and designing DT system. Additionally, we present the use case of digital patient and automatic remote surgery to give a detail explanation and demonstrate the efficacy of the architecture reference model: distributed DTs can analyze the current and historical medical data to plan, practice, and perform remote surgery. From feedback loop and machine learning module, DTs can gradually learn how to perform operations and make the operation automatic. This architecture reference model can works as a high level template to facilitate designing and integrating DTs across different systems, platforms, and domains.

First Page


Last Page




Publication Date



AI, architecture, cloud, digital patient, digital twin, federated machine learning, healthcare, IoT, IoT edge, machine learning, medical, multimedia, PaaS, remote surgery, robot


IR conditions: non-described