The human heart is our most vital organ, providing blood to all other organs in our body. Certain diseases, like elevated blood pressure, can negatively affect the hearts’ function and eventually lead to early death. In clinical practice, doctors choose the most suitable treatment based on an assessment of most invasively measurable indicators of the patients' physiological state, called biomarkers. This may rule out certain patients from therapy because of procedural risks and costs of gathering biomarkers and cannot be used easily for monitoring disease progression.

Non-invasive data collected during standard-of-care, like CT and MRI imaging data, provide information that could be used for improving assessment of clinical biomarkers. In the field of computational science, cardiac digital twins, a virtual replica of an individual’s heart have emerged as a promising technology that can harness more clinical data for improving our knowledge of the human heart and confidence in simulations thereof as they are based solely on the governing physics and physiology that drives a patient’s hearts function.

In this project, we want to unlock the full potential of cardiac digital twins by applying this concept not only to one individual but databases of virtual hearts. With the help of cutting-edge machine learning technologies, we will make it possible to study effects of varying patient anatomy as well as infer clinical biomarkers without the necessity for invasive measurements. Eventually, this holds the possibility to personalize treatments for patients and find the most suitable therapy for everyone.

Podcast Recommended - AirCampus (aircampus-graz.at)
ResearchGate

Project funding

Funding: EUR 522,350