New tech allows smartphones, smart speakers to detect cardiac arrest and call for help
Nearly 500,000 Americans die each year from cardiac arrest, a condition where the heart abruptly stops beating. People experiencing cardiac arrest will suddenly become unresponsive and either stop breathing or gasp for air, a sign known as "agonal breathing". Immediate CPR can double or triple someone's chances of survival, but that requires prompt help. Researchers in The University of Washington have developed a new tool — an app or an algorithm for smart speakers and smartphones — that allows it to detect the signature sounds of cardiac arrest and call for help.
Instead of listening for a word such as "Alexa", the smart device would listen for the sounds associated with a cardiac arrest and immediately connect victims to CPR support or Emergency Medical Services (EMS). The tool developed by the research team can detect agonal breathing 97% of the times when the smart device was placed up to six meters away from a speaker generating the sounds, state the findings published in Nature journal, npj Digital Medicine.
"During a cardiac arrest, victims will stop breathing normally, and in half the cases, they will exhibit a symptom known as agonal breathing, a type of disordered gasping sound. Given the proliferation of smart speakers and smartphones, we developed a non-contact technology on these devices that works by continuously and passively monitoring the room to identify agonal breathing and potentially call for CPR help by alerting emergency medical services or volunteer responders in the immediate area. Immediate CPR is critical as it can double or triple someone's chance of survival. And then if there's no response, the device can automatically call 911. Our technology can be thought of a smart speaker 'skill' that can be activated on your phone. Instead of listening for a wake word such as 'Alexa', the smart device would instead listen for these sound signatures," Justin Chan, a doctoral student at Paul G Allen School of Computer Science and Engineering, University of Washington, and first author of the study, tells MEA Worldwide (MEAWW).
Agonal breathing happens when a patient experiences very low oxygen levels. The research team said that according to 911 call data, agonal breathing is present for about 50% of people who experience cardiac arrest, and patients who take agonal breaths often have a better chance of surviving. However, said the research team, cardiac arrest frequently occurs in the privacy of someone's home or the patient's bedroom, where no one is likely to be around or awake to respond swiftly and provide help immediately.
This new tool developed by the researchers will be able to monitor people for cardiac arrest while they are asleep without touching them. They say that one could consider it as a new skill for a smart speaker — like Google Home and Amazon Alexa — or a smartphone that lets the device detect the gasping sound of agonal breathing and call for help. "The widespread adoption of smartphones and smart speakers (projected to be in 75% of US households by 2020) presents a unique opportunity to identify this audible biomarker and connect unwitnessed cardiac arrest victims to emergency medical services or others who can administer cardiopulmonary resuscitation (CPR)," the paper states.
For the study, the team collected sounds of agonal breathing from real 911 calls to Seattle's emergency medical services. As cardiac arrest patients are often unconscious, bystanders recorded the agonal breathing sounds by putting their phones up to the patient's mouth so that the dispatcher could determine whether the patient needed immediate CPR. The team collected 162 calls between 2009 and 2017 and extracted 2.5 seconds of audio at the start of each agonal breath to come up with a total of 236 clips. The team captured the recordings on different smart devices — an Amazon Alexa, an iPhone 5s and a Samsung Galaxy S4 — and used various machine learning techniques to boost the dataset to 7,316 positive clips. "Positive clips refer to clips containing agonal breathing," explains Chan.
The researchers also played these examples at different distances to simulate what it would sound like if the patient were at different places in the bedroom. They added different interfering sounds such as sounds of cats and dogs, cars honking, air conditioning, things that one might typically hear at home.
On average, the proof-of-concept tool detected agonal breathing 97% of the times, from up to 20 feet or six meters away, say the researchers. "Cardiac arrests are a common way for people to die, and right now, many of them can go unwitnessed. Part of what makes this technology so compelling is that it could help us catch more patients in time for them to be treated," co-author Dr. Jacob Sunshine, an assistant professor of anesthesiology and pain medicine at the University of Washington School of Medicine, says in a statement.
The researchers also tested the algorithm to make sure that it would not accidentally classify a different type of breathing, like snoring, as agonal breathing.
The researchers plan to commercialize this technology through a University of Washington spinout, Sound Life Sciences, Inc. "Getting more data from 911 calls across the country will be needed to improve the technology's accuracy. These datasets will be nice to have before the software goes on the market. The software is being licensed by a startup called Sound Life Sciences. They plan to make it available in the next year or so," Chan tells MEAWW.