Coronary artery disease (CAD) is the most common type of heart disease, and a leading cause of death for both men and women in Canada. It’s caused by plaque buildup in the walls of the arteries that supply blood to the heart and can lead to serious problems, including chest pain and heart attacks.
Physicians have two tests available to identify blockages in the arteries: coronary angiography (ICA), which involves inserting a catheter into the heart artery; and coronary computed tomography angiography (CCTA), which uses non-invasive imaging techniques.
Neither test is perfect for all patients. While CCTA is less invasive, exposes patients to less radiation and is a less expensive test than ICA, it may overestimate the severity of blockages. If patients are found to have significant blockages on CCTA or have ongoing symptoms, they may then have to undergo an ICA test. Deciding which initial test patients should receive isn’t always clear.
Dr. Robert Miller, MD, a clinical associate professor in the Cumming School of Medicine, is tackling this problem. He recently received a $500,000 Alberta Innovate Grant to develop artificial intelligence (AI)-powered software to help physicians direct patients to the most appropriate test.
It’s the first time AI has been used to assist physicians with making this decision.
“It’s an exciting project that will help us better select patients for appropriate testing,” says Miller. “Our goal is to develop software that can be used clinically to help our patients.”
The software will leverage de-identified patient information — accessed within the Libin Cardiovascular Institute’s Precision Medicine Initiative’s data pipeline — to predict the most appropriate test. Miller and his team will evaluate the performance of the software in Calgary and Edmonton.
Miller says the project has the potential to cut down on wait times for tests and to save health-care system resources, all while ensuring patients receive safe and efficient care.
“It’s really exciting to be able to address a clinical need,” says Miller. “This research is relevant and can improve patient outcomes right away. It’s very rewarding.”