Cardiovascular diseases: New computer model improves therapy

In order to be able to rule out such interventions in advance, Plank has developed a computer model together with the mathematicians Gundolf Haase and Kristian Bredies from the University of Graz and computer scientist Thomas Pock from the Institute of Computer Vision and Representation at Graz University of Technology, respectively, with which doctors can pre-simulate the optimal therapy and dramatically improve the success of treatment.

The researchers use diagnostic data from MRI, ECG and other heart examinations of the person to be treated. Imaging algorithms put together a digital image of the patient's heart from this data material. This customized model ultimately provides a wealth of information that helps to understand the individual clinical picture and to run through various therapeutic scenarios.
Thomas Pock explains the challenge behind it: "To simulate such a heartbeat in the computer, you have to calculate millions of variables. This requires complex mathematical procedures, special algorithms and special hardware that can perform billions of computing actions per second."

The developed method is so sophisticated and automated that anatomically correct digital twins of patient hearts can already be routinely produced in a clinical setting. In a next step, the researchers want to further improve the technology and enable fully automatic adjustment of all functional aspects of the heartbeat. "This requires further efforts in basic research, especially in those areas of machine learning and artificial intelligence (AI) that allow a high degree of personalization," explains Pock.

A very promising approach is based on the latest AI methods for optimal control and focuses on the wave propagation in the heart, which is controlled by the alignment of the heart muscle fibres. The consortium wants to implement this approach in cooperation with Cardiocentro Ticinio (centre for computer-assisted cardiology, Lugano) in a new research project and try to incorporate the "control elements" into the model using machine learning techniques in such a way that the simulated heartbeat comes as close as possible to the real heartbeat.

The first clinical validation studies are in preparation for 2021 in cooperation with Daniel Scherr from the Division of Cardiology at the Medical University of Graz. Plank and Pock assume that clinically usable prototypes of a fully automated digital twin heart can be tested as early as 2022. The simulation technology on which the method is based is already being distributed by the Graz-based start-up NumeriCor and is used by leading medical technology companies in the R&D sector.

The participating researchers are part of the inter-university research network BioTechMed-Graz. This is a shared initiative of the Medical University of Graz, Graz University of Technology and the University of Graz in the field of biomedical sciences.

This research topic is anchored in the Field of Expertise “Information, Communication & Computing”, one of the five strategic research foci at TU Graz.

COMPAMED-tradefair.com; Source: Graz University of Technology