A method of monitoring cardiac health of a human (user) includes measuring back portion cardiac data using one or more sensors that are all in contact with a posterior of a human torso. The method further includes detecting a cardiac anomaly using machine learning based on the back portion cardiac data. In one or more examples, the method further includes converting the back portion cardiac data to front portion cardiac data using a translation model, wherein the front portion cardiac data is used for detecting the cardiac anomaly. In one or more examples, the one or more sensors are positioned using a panel that is incorporated into a garment that is in contact with the human torso. The panel can be detachable from the garment in one or more examples.
