This software provides the processing for a non-contact system that remotely estimates the heart rate and respiration rate of individuals as they carry on daily activities, and also enables detection of heart and respiration rate through walls.
Prior attempts have been made using microwave Doppler radar techniques to estimate heart rate and respiration remotely. However, such techniques lacked advanced signal processing to effectively “deconvolve” the heart rate and respiration signals from human body motions from the Doppler signal reflections. Therefore, competing approaches have only been successful in estimating heart rate or respiration when the individual is completely stationary.
This software remotely estimates human vital signs, such as heart rate and respiration rate, from microwave reflections from humans using high-frequency, narrow-band microwave (18 to 30 GHz). The signal processing algorithms implemented in this system allow estimation of heart rate and respiration rate from microwave reflection from the torso of an individual. It uses novel signal processing and advanced machine learning techniques to detect unique heart signatures, even in the presence of natural body motions such as moderate breathing or fidgeting.
This technique can be used to monitor the vital signs of astronauts on the International Space Station, during long space missions to the Moon or Mars, and during spacewalks, when using contact electrocardiogram and respiration belts is cumbersome and interferes with the activities of astronauts.
This work was done by Ashit Talukder and Steven P. Monacos of Caltech for NASA’s Jet Propulsion Laboratory.