Major grant to strengthen cardiac diagnostics with artificial intelligence

Clinical Associate Professor and Consultant Simon Winther from the Department of Clinical Medicine and Gødstrup Regional Hospital has received DKK 34 for a major Nordic research project on artificial intelligence in cardiac diagnostics.

Simon Winther and Samuel E. Schmidt are leading the project Nordic Collaboration on AI and Cardiac Imaging, which aims to develop publicly available AI models for the analysis of cardiac CT scans.
Simon Winther and Samuel E. Schmidt are leading the project Nordic Collaboration on AI and Cardiac Imaging, which aims to develop publicly available AI models for the analysis of cardiac CT scans. Photo: Søren Braad Andersen, Gødstrup Regional Hospital.

The aim of the project, which is supported by the Novo Nordisk Foundation, is to strengthen the diagnosis of cardiovascular disease by building a comprehensive Nordic database of cardiac CT scans and establishing a shared platform where researchers and clinicians can develop and test AI models across national borders.

“This project is a fantastic opportunity to bring together a cross-disciplinary team that can develop high-quality AI models, which we will thoroughly test and compare with other leading methods. The models will be open source, allowing other researchers to build upon them. This could further accelerate development. We hope the project will promote the use of AI models in everyday clinical practice and ultimately benefit both patients and the healthcare system,” says Simon Winther.

The dataset will include scans from thousands of individuals in Denmark, Norway, and Sweden, ranging from healthy individuals to patients with severe atherosclerosis. This will enable researchers to analyse disease progression over time. Advanced calculations of blood flow in the coronary arteries of the heart will be performed using the Danish supercomputer Gefion.

The researchers will rigorously validate the AI models before they are implemented and used to support decision-making in clinical practice. The project will only involve non-identifiable data, and only the source code for the developed models will be shared. All data processing will take place in secure research environments in accordance with applicable regulations and approvals.

The project is a collaboration with Professor Samuel E. Schmidt from Aalborg University.

The text is based on press material from Gødstrup Regional Hospital.

Contact

Clinical Associate Professor and Consultant Simon Winther
Aarhus University, Department of Clinical Medicine and
Gødstrup Regional Hospital, Department of Cardiology
Phone: +45 25 12 10 10
Email: swinther@clin.au.dk