TEHRAN, Jun 20 -Researchers at the University of Washington developed a tool for smart speakers such as Amazon's Alexa and Google's Home that can monitor people for cardiac arrest while they're sleeping -- and call for help if needed.
TEHRAN, Young Journalists Club (YJC) -Smart speakers already help people cook, operate other devices in their homes, and listen to music, among other tasks, but researchers at the University of Washington think they could also help save lives.
One sign that a person is suffering cardiac arrest is if they gasp for air, which can help bystanders know whether to perform CPR.
Researchers at UW developed a tool to detect the gasp -- called agonal breathing -- that is specific to cardiac arrest, finding that it was correctly detected 97 percent of the time by Google Home, Alexa and other smart devices, according to a proof-of-concept study published Wednesday in npj Digital Medicine.
The researchers who developed the digital tool hope it's use could ultimately give first responders enough time to treat cardiac arrest patients.
"We envision a contactless system that works by continuously and passively monitoring the bedroom for an agonal breathing event, and alerts anyone nearby to come provide CPR. And then if there's no response, the device can automatically call 911," Shyam Gollakota, an associate professor at the University of Washington and study corresponding author, said in a news release.
Agonal breathing, which affects 50 percent of cardiac arrest sufferers, gives people a better chance of living after an event, the researchers say.
When a person falls unconscious from cardiac arrest, dispatchers instruct bystanders to put their phones to the person's mouth to pick up any sound of agonal breathing. If so, the dispatcher tells the bystander to perform CPR.
For the study, the researchers collected 162 calls to 911 from the Seattle Emergency Medical Services between 2009 and 2017. They clipped about 2.5 seconds of sound from each call when the agonal breathing began. In all, the researchers used Amazon Alexa, an iPhone 5s and a Samsung Galaxy S4 devices to gather 236 clips.
Using multiple machine learning techniques, they increased that number to more than 7,300 clips of agonal breathing.
Once they fix its glitches, the researchers want roll the tool out to market, which they say could run as an app, or as a skill that Alexa runs passively in the background.
"Cardiac arrests are a very common way for people to die, and right now many of them can go unwitnessed," Jacob Sunshine, an assistant professor of anesthesiology and pain medicine at the University of Washington and study corresponding author. "Part of what makes this technology so compelling is that it could help us catch more patients in time for them to be treated."
Source:upi