Grant: Project Grant

The electrocardiogram (ECG) is a common medical test that artificial intelligence (AI) technology now aids in analyzing more effectively. Previously, our team used ~2 million ECG collected from emergency departments (ED) and hospitals to develop AI algorithms capable of predicting current health conditions and future adverse events. However, patients who present at ED or hospitals are likely sicker than individuals who get an ECG in the community setting. The ability of community-obtained ECG to predict heart issues requires investigation. Essentially, these intelligent programs may be able to inform us about possible future health problems. In this study, our main goals are to improve AI algorithms that use ECG to diagnose heart issues and predict future problems. We will see if the algorithms we have developed in acute care settings work well for community settings. Additionally, we will create new algorithms to identify individuals at a higher risk of future adverse events. In our past CIHR-funded project, we connected ECG collected in ED and hospitals with bigger databases that contain health information on doctor visits, medications, and lab tests. These were linked to vital statistics death registry to assess mortality outcomes. In the current study, we intend to establish similar links for community-obtained ECG. We will use traditional machine learning and cutting-edge deep learning techniques to develop AI algorithms. The research team includes specialists in cardiology, AI, and epidemiology from Canadian VIGOUR Centre and Alberta Machine Intelligence Institute (UoA), and Centre for Health Informatics (UoC). We are working with patient partners as we design the project. Our project resonates with Canada's pursuit of establishing a learning health system-an approach centered on continuous health data collection and analysis. The study could lead to personalized health approaches focusing on wellness, disease prevention, and tailored treatments.