Researchers at the University of Cambridge, UK, have used AI to create Digital Twins of people’s body, the digital twin of your body. The goal is to help monitoring your wellbeing and detect the insurgence of sickness before you perceive any symptoms. Having the possibility to become aware that something is wrong at a very early stage can make a huge difference on the outcome, be it a Covid-19 infection or a cancer.
You can find the detailed explanation of their approach and the first evaluation of results in their paper “Graph representation forecasting of Patient’s medical conditions: Towards a Digital Twin“.
There have been, and there are, several applications of digital twin technology in the healthcare domain, including using digital twins to model a human body for simulation (as an example in pharma to speed up drugs development), for monitoring … and of course AI is being used in the analyses of data but to my knowledge this is the first time that AI is used to create a human body digital twin.
The stated objective of their research is to foster preventative medicine providing personalised treatment plans to patients. To do that they apply a number of AI technologies, including GNN – Graph Neural Network- and GAN – Generative Adversarial Network- feeding a mathematical model, embedded in the Digital Twin, of current and future pathophysiological conditions.
The Digital Twin model has been tested on simulated clinical case studies with data at organ, tissue and cellular level. AI was used to asses current conditions based on the available data and to forecast probable evolution.
This research has not reached a clinical trial stage but represents a proof of concept in the use of Digital Twins automatically generated using AI.