I am a pediatric cardiologist and have cared for children with heart disease for the past three decades. In addition, I have an educational background in business and finance as well as healthcare administration and global health – I gained a Masters Degree in Public Health from UCLA and taught Global Health there after I completed the program.
“Be yourself. An original is always worth more than a copy.”
Suzy Kassem, writer and visionary
This review article from Journal of Personalized Medicine is a reasonably good overview for any interested clinician to begin to understand the relatively new realm of digital twins and the relevance to artificial intelligence. A digital twin is “a virtual model of a physical entity, with dynamic, bidirectional links between the physical entity and its corresponding twin in the digital domain.” This new virtual model leverages current technologies such as smart sensors, edge computing, extended reality, and artificial intelligence in order to iteratively model, test, and optimize the physical object (person in the case of healthcare) in the virtual space until a certain performance is met.
The digital twin technology can therefore create a new research paradigm in the form of a personalized in silico methodology of investigation. With the recent emphasis on real world data and evidence generation and the ongoing debate of the true value of traditional randomized controlled trials, a digital twin may be the eventual ideal real-world, real-time research tool. The digital twin feature of predicting performance is becoming an increasingly more important one with the integration of artificial intelligence. Another key feature of digital twins is dynamic bidirectional mapping, a feature that renders digital twins far beyond a simulation model of the physical entity in the digital domain.
The authors further describe the various different types of digital twins (whole body, organ system and/or disease states, or cellular components) and even composite digital twins that incorporate two or more types. In addition, aggregates of digital twins can represent a population and healthcare institutions can have a digital twin of the entire organization (DTO). Digital twins can have identical copies called digital twin instances that can be used to perform in silico comparative testing. Digital twins are being explored for drug discovery, precision cardiology, and even precision public health coupled with a smart healthy city.
As with other emerging technologies such as artificial intelligence, digital twins will have ethical and regulatory challenges that will need to be solved. The coupling of digital twins with artificial intelligence (“intelligent digital twins”) has the potential to change the future research paradigm.