In our modern way of life, WiFi connection has become almost as essential as water, electricity or food. But if the primary function of this technology is to transmit data from a router to our smartphones, our PCs and our televisions, scientists regularly look into the misuse of this technology. And during the COVID-19 pandemic, researchers at the National Institute of Standards and Technology, a US government agency, worked on a way to use WiFi waves to detect respiratory illnesses.
“As everyone’s world was turned upside down, several of us at NIST were thinking about what we could do to help”, says Jason Coder, a researcher at the agency, quoted by New Atlas. He adds that as they did not have time to create a new device, they wondered how they could exploit an existing product already in homes. And this is how the idea of diverting Wifi for the detection of respiratory diseases came about.
Your breathing leaves a trace on the waves of Wifi
The principle is quite simple to explain: Wifi, to transmit information from one point to another, crosses obstacles. Among these obstacles are humans. When the waves pass through a person, they are altered. And it is these alterations that the researchers studied in order to determine whether Wifi could be used to detect respiratory disease.
The idea was tested using a dummy simulating the breathing of a human being, in an experiment room called an anechoic chamber, which absorbs waves. A Wifi router and receiver device have also been placed in the room. And during the experiment, the dummy simulated asthma, chronic obstructive pulmonary disease, and other conditions where breathing is abnormal. At the same time, a device records the variations on the alterations of the waves.
© NIST
A promising technology, but…
From these experiments, the researchers then developed an artificial intelligence which, according to them, was able to correctly classify the type of breathing, with an accuracy of 99.54%. In other words, by analyzing the WiFi waves, the algorithm is able to tell if the model is simulating an asthma attack, or another respiratory problem.
Unfortunately (or not), the research stopped there. And we can assume that in real conditions, without the experimental chamber, and on a hearth with real humans, this algorithm could have more difficulties.
Nevertheless, it is a proof of concept, which could pave the way for more advanced research. And possibly, in the future, there could be a mobile application which, by studying WiFi signals, will be able to warn you if you have a health problem.
