Title:	Digital Twin
DNr:	LiU-2019-18
Project Type:	LiU Compute
Principal Investigator:	Tilda Herrgårdh <tilda.herrgardh@liu.se>
Affiliation:	Linköpings universitet
Duration:	2019-05-16 – 2024-06-01
Classification:	10610
Keywords:

Abstract

Systems biology emerged as a way of understanding biology at a system level, e.g. organ or some other biological system. However, most diseases don’t act solely on an organ level, but on a whole-body level. That is, to understand most diseases and medical problems, a more holistic understanding is needed. We currently have several mechanistic models describing the dynamics of different biological systems at both cellular level and whole-body level, on a time scale from seconds to years. More specifically, we have models describing insulin signalling in adipose tissue, glucose homeostasis, weight change over years, liver uptake and inflammation, brain metabolism, and blood flow. We are currently working on connecting these models with each other. The connected models can then be personalized by training them to individuals’ data and used to simulate how different biomarkers would change over time given different scenarios. We also want to connect these mechanistic models, describing how different biomarkers change over time, with machine learning models to be able to incorporate different kinds of data and to calculate risk predictions for different diseases, given both current data and the predictive simulations from the mechanistic part. The resulting hybrid model can be used as decision support in a medical setting, for pharmaceutical development, or by individuals wanting to understand their body better.