Noninvasive System Watches Over Respiratory Patients

The researchers developed a maximum likelihood estimator (MLE) to estimate breathing parameters, including breathing rate, using all of the measured links’ Received Signal Strength (RSS) data feeds simultaneously.

The researchers then developed detection algorithms based on those estimated parameters, and an experimental test bed and procedure to validate BreathTaking. Using extensive data collected with a person lying in a hospital bed, they found that the breathing rate could be estimated within 0.1-0.4 breaths per minute error, using 30 seconds of measurements. The researchers demonstrated that they could reliably distinguish between breathing and its absence using 15 seconds of RSS data (in patch antenna experiments) and using 30 seconds of data (in dipole antenna experiments), without false alarm or missed detection. The study was submitted on September 18, 2011, to IEEE Transactions on Mobile Computing.

“The patient or the baby doesn't have to be connected to tubes or wired to other sensors, so they can be more comfortable while sleeping. If you're wired up, you're going to have more trouble sleeping, which is going to make your recovery in the hospital worse,” said senior author Neal Patwari, PhD, an assistant professor of electrical engineering. “If BreathTaking is to be used for medical purposes, extensive evaluation on many people, and in many settings, must be performed.”

The system design is based on the observation that when a person is standing on or near the line-of-sight of a static radio link, the RSS can be affected simply by the person’s inhaling and exhaling. BreathTaking, in fact, does not provide a direct measure of breathing; it simply makes an observation about the presence (or absence) of a strong frequency component in the measured RSS in the human breathing range.

Related Links:
University of Utah