Multimodal characterization and classification of bio-signals to predict cardiac arrest

Sudden cardiac arrest (SCA), due to abrupt disruption of cardiac function, is a major health problem globally. SCA can happen to anyone at any age who may or may not have been diagnosed with heart disease. SCA has a poor survival rate of about 10 percent, with an estimated 35,000 deaths in Canada annually. With an increasing rate of cases (16 percent from 2017 to 2020), SCA remains a major public health issue in British Columbia. The most effective strategy to improve survival is to achieve rapid SCA recognition, given that for every minute without cardiopulmonary resuscitation (CPR) survival rates drop by 10 percent. Wearable devices may play a major role in decreasing SCA mortality, providing real-time cardiac information for early SCA detection. My aim is to develop a wearable SCA device with embedded sensors, and use their real-time physiological data combined with artificial intelligence algorithms, to make an accurate SCA detection system. This SCA detection system will be designed to identify SCA and alert Emergency Medical Services with the individual’s location (via GPS), enabling them to provide life-saving interventions in a timely manner.