The STRATIF-AI project, led by Gunnar Cedersund at Linköping University, is funded with 65 million SEK over four years from the European Horizon (EU Commission). The project aims to utilize digital twins to enhance preventive measures, treatment, and rehabilitation for stroke patients. By creating a digital copy of a person's body, known as a digital twin, advanced technology can be used to optimize healthcare and assist people in living healthier lives. Although the project primarily focuses on stroke research, we believe that digital twins have a wide range of future applications.
Accessible healthcare through digital twins and continuous stratification
Today, there is a wealth of patient data that is not consistently utilized. This is due to data being stored in various formats and originating from different sources, leading to healthcare providers not always having access to all the necessary patient information. To address this, STRATIF-AI introduces a new technology - continuous stratification, using our new STRATIF-AI platform.
Within the STRATIF-AI platform, all patient data is continuously stored in a Personal Data Vault, which the patient can control themselves. Essentially, with this data the twin can continuously evolve and provide a dynamic and real-time representation of the patient's health status.
Useful apps for stroke patients
The data within the digital twin can be linked to various apps that can support the patient throughout their healthcare journey. For instance, an app can simulate how a patient might respond to changes in their diet, exercise, or medication. We can observe these reactions at all levels, from the smallest cells in the body to the overall functioning of the body, and this can be done over various timeframes, ranging from seconds to several years.
Digital twin models for stroke patients
Gunnar Cedersund has been working on simulating human organs for over 20 years, somewhat like assembling a digital puzzle of the human body. This is achieved by combining machine learning and mechanistic models. Using machine learning, a comprehensive data model is constructed using research data and knowledge from hospitals, universities, and biobanks to create risk models for stroke.
In this project, we will, for the first time, employ this cutting-edge technology to connect a series of apps that collectively cover the entire patient journey for stroke patients. We will conduct six new clinical studies involving eight partner hospitals in the EU to enhance and validate our models and demonstrate how the same digital twin can accompany a patient through all phases of stroke: from prevention to acute treatment and rehabilitation.
