Difficulty in breathing is one of many important warning signs of a pending medical emergency. Now, researchers at the National Institute of Standards and Technology (NIST) and other federal institutions are exploring AI-based sensing applications that may help quickly identify individuals experiencing health-related breathing issues.
According to an article posted to the NIST website, researchers at NIST and the Office of Science and Engineering Labs (OSEL) at the U.S. Food and Drug Administration (FDA) are exploring the use of Wi-Fi routers to measure the breathing rate of a person in a room. Their work may soon pave the way for tools that can more quickly identify an abnormal breathing pattern and result in prompt action to head off a serious health condition.
In brief, the researchers modified Wi-Fi firmware to increase the stream of channel state information (CSI) traveling between a client device and the router. They then developed a deep learning algorithm, code-named BreatheSmart, to analyze how an individual body moves during breathing, and to recognize breathing patterns that may indicate breathing problems. They found that BreatheSmart was remarkably effective in classifying a variety of respiratory breathing patterns, with a 99.54% success rate.