Digital Twins for Health A Scoping Review

by Katsoulakis et al

Annotated Link

Definition & Character of digital twin:

  • A virtual representation of objects

  • Faithfully mirror the real-world system in real-time, analyze its behavior and provide predictive insights using advanced simulation, ML and reasoning.

Types of Digital Twin

  • Static twin: A digital replica of a physical system

  • Mirror (functional) twin: A static twin with dynamic behavior capabilities

  • Shadow (self-adaptive) twin: A functional twin with the capacity to acquire real-time data and update the model

  • Intelligent (extended/cognitive/physical avatars) twin: Self-adaptive twin with a degree of artificial intelligence that has autonomy with learning, reasoning, knowledge, and acting capabilities.

Different types of DTs for healthcare:

  • Provided a somewhat OK overview of different applications.

(Thoughts): LLM vs. ML models

  • ML models have limited functions. A prediction model can only do prediction. However, if we replace the inference part with an LLM, the final product will be moved into the realm of a digital twin.

  • Digital phenotype: patterns in human body and behavior. (Input to DT construction).

Last updated