Concepts and applications of digital twins in healthcare and medicine

Abstract

The digital twin (DT) is a concept widely used in industry to create digital replicas of physical objects or systems. The dynamic, bi-directional link between the physical entity and its digital counterpart enables a real-time update of the digital entity. It can predict perturbations related to the physical object's function. The obvious applications of DTs in healthcare and medicine are extremely attractive prospects that have the potential to revolutionize patient diagnosis and treatment. However, challenges including technical obstacles, biological heterogeneity, and ethical considerations make it difficult to achieve the desired goal. Advances in multi-modal deep learning methods, embodied AI agents, and the metaverse may mitigate some difficulties. Here, we discuss the basic concepts underlying DTs, the requirements for implementing DTs in medicine, and their current and potential healthcare uses. We also provide our perspective on five hallmarks for a healthcare DT system to advance research in this field.

Figure 1

Basics of DTs A virtual entity, a physical entity, and an input data flow for real-time collection and monitoring of the physical entity’s state or physiological functions, along with an output data flow for real-time interaction and communication, such as transmitting diagnosis and treatment solutions.

Figure 2

Building with AI and metaverse (A) Building DTs with LLMs. (B) Combining embodied AI with LLM-powered DTs to construct AI agents. (C) Metaverse provides a shared space for physical and virtual entities to communicate regarding patient care.

Figure 3

Hallmarks of the DT platform Any healthcare DT should include basic physical-virtual two-way communication, a metaverse of representative data, embodied AI agents based on LLM interfaces, reliable learning and prediction of multi-modal data, real-time patient monitoring, secure data storage, access to patient data, and adherence to ethical standards.

Figure 4

Representative data repository as a metaverse (1) The DT platform integrates extensive multi-modal biological omics and medical data from patients, generating algorithms for individualized guidance in prevention, risk assessment, and therapies. (2) The platform uses comprehensive patient input to match their virtual counterpart in the deeply phenotyped DT database, providing personalized treatment and prevention recommendations.

Figure 5

Development of the DT platform Static twins serve as the starting point, where physical entities are digitized, enabling periodic updates to their virtual counterparts. By integrating temporal or progressive information, progressive twins can reflect the evolution of the physical entity and reliably forecast future state transitions. With the development of a closed-loop, iterative improvement framework, operational twins enable real-time interaction between physical and virtual entities. This facilitates both a deeper understanding of biological phenomena and the achievement of specific design objectives in biology and healthcare. In the final stage, the digitized physical and virtual worlds merge, representing the highest level of physical-virtual co-existence, or autonomous twins. Autonomous virtual entities continuously generate information and knowledge for their associated physical entities within the DT platform.