STTR Phase I: Using Mechanical Power for Cardiac Risk Stratification and Rehabilitation

The broader impact/commercial potential of this Small Business Technology Transfer (STTR) Phase I project is to increase access to and compliance with cardiac rehabilitation program and reduce the cost of avoidable hospital readmission, both by predicting which cardiac patients are at risk and by providing patients with simple but accurate guidance, allowing them to precisely control exercise intensity and volume during rehabilitation. It also has the potential to reduce the mortality rates of cardiac patients and provide doctors with a large amount of accurate information on the relationship between exercise duration, exercise intensity, and changes in cardiac patient conditions during recovery, supporting optimal design of rehabilitation program design. The key idea enabling these outcomes is using wearable sensors to measure the mechanical power outputs of cardiac patients and using these power measurements to estimate an established high-quality predictor for cardiac patient hospital readmission and mortality rates. The size of the market is substantial, with roughly one million Americans suffering from heart attacks per year. The proposed project will determine whether total body mechanical power output can be used to predict metabolic energy expenditure, using a ground truth based on volume of oxygen consumption per unit time. This metric, VO2 is a heavily used indicator of cardiac health for patients who have experienced heart attacks. Ambulatory mechanical power output has been used by athletes to track and provide feedback on physical condition, but it remains unclear whether similar applications are feasible for cardiac patients, due to their differing physical conditions and sensor use preferences. In addition to validating the use of mechanical power to estimate VO2 the project will also evaluate different sensor use and wearability conditions appropriate for cardiac patients, and make innovations in signal processing algorithm, sensor design, and usability required by these conditions.