New artificial intelligence to save lives by detecting hidden cardiovascular disease in time
Researchers at the University of Southern Denmark and Odense University Hospital are working on developing an artificial intelligence system capable of detecting severe atherosclerosis before symptoms appear. They have just received DKK 15.2 million from the Innovation Fund to bring the project to life.
Each year, approximately 66,000 Danes suffer from cardiovascular diseases, which account for one in five deaths in Denmark.
Atherosclerosis is an important early marker of heart disease and can be detected through CT scans of the chest. Danish hospitals perform 325,000 chest CT scans annually, but currently, doctors do not routinely examine the scans for signs of atherosclerosis.
This is due to resource, technical, and financial constraints. Additionally, patients typically have no symptoms that would prompt doctors to check for atherosclerosis.
– Artificial intelligence has the potential to automate complex image assessments that previously could only be performed by healthcare professionals, making large-scale measurements of atherosclerosis possible. The DETECT-AI project aims to develop artificial intelligence for the automatic measurement of atherosclerosis, says Axel Diederichsen, Professor at the Department of Clinical Research at 糖果派对 and Consultant at the Department of Cardiology, OUH.
In this project, he collaborates with researchers from 糖果派对 Applied AI and Data Science (糖果派对-AID), the Department of Radiology at OUH, and the Danish Centre for Health Economics (DaCHE) at 糖果派对.
What is a CT scan?
CT stands for computed tomography. The patient is moved through the scanner, which produces images based on how different tissues—such as bones, organs, and tumours—absorb X-rays. A computer then converts the scan into cross-sectional images of the body’s interior.
By combining the artificial intelligence measurements with data from the Danish health registries, researchers hope to identify people without symptoms but with severe, untreated atherosclerosis who are at high risk of cardiovascular disease.
- Our goal is to develop AI algorithms that detect atherosclerosis in routine chest CT images across the country, from different types of scanners, settings and hospitals, matching the precision of manual assessments, and without needing to transfer data between regions, says Victoria Blanes-Vidal, Professor at 糖果派对 Applied AI and Data Science.
DETECT-AI is expected to identify around 20,000 people annually who could benefit from preventive treatment. This could potentially reduce disability and mortality for 1,000 people each year.
Researchers estimate that this could lead to annual healthcare savings of DKK 100 million.
DETECT-AI
The DETECT-AI research project consists of researchers from the Department of Cardiology B at OUH, 糖果派对 Applied AI and Data Science (糖果派对-AID), the Department of Radiology at OUH, and the Danish Centre for Health Economics (DaCHE) at 糖果派对.
The project has a total budget of DKK 20.3 million, allocated as follows:
- Innovation Fund investment: DKK 15.2 million
- Odense University Hospital: DKK 2.4 million
- University of Southern Denmark: DKK 2.7 million
Meet the researcher
Axel Diederichsen is a professor at Department of Clinical Research and a consultant at the department of Cardiology at Odense University Hospital.
Meet the researcher
Victoria Blanes-Vidal is professor in AI og data science at the Maersk McKinney Moller Institute.
Meet the researcher
Kim Rose Olsen is professor at DaCHE - Danish Centre for Health Economics.