Multi-Modal Wearables with Machine Learning Algorithms for Continuous Cardiac Monitoring: Implications for Timely Detection of Sudden Cardiac Arrest

Sudden cardiac arrest (SCA) is a major health issue where the heart unexpectedly stops beating, and blood no longer circulates through the body. In Canada, there are about 60,000 cases of SCA each year. Quick action, such as starting cardiopulmonary resuscitation (CPR), is crucial. The longer it takes to start CPR, the less likely a person is to survive. Unfortunately, most SCAs happen in places where no one is around to help. Therefore, one of the biggest challenges in improving survival rates is the delay in recognizing that a cardiac arrest has occurred. Consumer wearable technologies have great potential to continuously monitor cardiac function and automatically recognize signs of disruption or loss of blood circulation at any time. If a problem is detected, these devices could automatically call emergency services for help. Our previous studies have shown that wearable sensors combined with artificial intelligence (AI) can recognize some signs and symptoms associated with a cardiac arrest, such as absence of a pulse. The overarching goal of the proposed project is to validate the performance of such a system for real-world SCA detection.