<p dir="ltr">HYPOTHESIS:</p><p dir="ltr">Machine learning models leveraging Mobile data, HRV, Hemodynamic data, Temporal Features and Patient Demographics can detect early physiological markers of Sudden Cardiac Arrest up to 72 hours in advance, with predictive accuracy (p < 0.05 by DeLong test) compared to mobile only data.</p>