For many people, Wi-Fi is a symbolic lifeline – new research from the National Institute of Standards and Technology (NIST) proves it can be a literal lifeline.
By modifying an off-the-shelf Wi-Fi router with firmware updates and a deep learning algorithm, researchers were able to detect breathing patterns indicative of respiratory failure in a medical manikin. The idea of trying to collect Wi-Fi signals to monitor people’s breathing patterns in their homes arose during the height of the COVID-19 pandemic. As everyone’s world was turned upside down, a few of us at NIST thought about what we could do to help said NIST researcher Jason Coder. We haven’t had time to develop a new device, so how can we use what we already have?
Read also : See what happens when you don’t lower the toilet seat?
The answer came by looking at the radio waves that enable communication between devices like cell phones or tablets and the routers they use to connect to the internet. During their round trips, these radio waves encounter obstacles such as furniture or people that slightly alter them.
By studying these changes, Coder, research associate Susanna Mosleh and her colleagues at the Office of Science and Engineering Labs (at the FDA’s Center for Devices and Radiological Health) believed they could detect subtle changes in the body of a person that would indicate breathing difficulties. -Fi was used to count people through walls and monitor sleep patterns.
To test the idea, the team placed a breathing dummy in a radio-absorbing room called an anechoic chamber. They also installed a commercial Wi-Fi router and receiver. As the dummy mimicked various breathing patterns, including those believed to indicate asthma, COPD, and abnormally slow and rapid breathing rates, radio frequency interference was recorded and data was transmitted approximately 10 times per second.
This led to the collection of a large amount of information which had to be analyzed to find out which wave disturbances corresponded to simulated breathing difficulties in the dummy. To sift through all of this, Mosleh created a deep learning algorithm that the team called Breathe smart. After setting up the equation and entering the data, it was found to be 99.54% efficient at classifying breathing patterns correctly.
The fact that the system can work with existing routers gives researchers hope that one day it could be deployed simply by using a smartphone app that provides a router firmware update. They also say their work creates a framework into which other types of surveillance algorithms can fit.
Of course, testing in a closed room with a medical dummy will be completely different from real-world applications where people move between furniture, pets, and each other, but the research is at least proof of concept for the system – it is already very promising!
Information about the research was published in the journal IEEEAccess.
➔ Follow us on Google News to stay up to date!
source: NIST | New Atlas
I’ve been dealing with Internet communication and e-marketing since 2005, I’m passionate about mobile devices and new technologies – and I don’t hesitate to use them.