Respiration and Cardiac Activity Sensing Using 3-D Cameras.

Remote vital sign monitoring at home with commercially viable hardware extends care from hospital wards to personal settings without overwhelming health care providers and resources. In this paper, we explore usage of Azure Kinect (Microsoft) to monitor respiration and cardiac activity of a subject. We propose a fusion sensing technique that ensures immunity to ambient lighting conditions and movements of the subject, two key limitations while employing cameras or radars respectively. The functionality of this process is enabled by the phenomenon of Photoplethysmography (PPG) and availability of commercial sensors that can leverage it. We investigate robustness of the proposed technique by conducting experiments in variable lighting conditions while the subject is in motion. A comparison is drawn to an RGB camera and the results are validated using a pulse oximeter, a contact PPG sensor.