Researchers develop first contactless cardiac arrest AI system for smart speakers

Almost 500,000 Americans die each year from cardiac arrest, when the heart suddenly stops beating.

People experiencing cardiac arrest will suddenly become unresponsive and unrestricted. Immediate CPR can double or triple someone's chance of survival, but that requires a bystander to be present.

Cardiac arrests often occur outside of the hospital and in the privacy of someone's home. Recent research suggests that one of the most common locations for an out-of-hospital cardiac arrest is in a patient's bedroom, where

Researchers at the University of Washington have developed a new tool to monitor people for cardiac arrest while they're asleep without touching them. A new skill for a smart speaker- like Google Home and alexandre Alexa-or smartphone lets the device detect the gasping sound of agonizing breathing and call for help. On average, the proof-of-concept tool, which has been developed using agonal breathing events. The findings are published June 19 in npj Digital Medicine.

"Said co-corresponding author Shyam Gollakota, an associate professor in the UW's Paul G. Allen School of Computer Science & Engineering . "We envision a contactless system that works continuously and continually monitors the chamber for an agonizing breathing event, and alerts us to provide CPR, and then, if there is no response, the device can automatically call 911."

Agonal breathing is present for about 50% of people who experience cardiac arrests, according to 911 call data, and patients who take agonal breaths often have a better chance of surviving.

Dr. Jacob Sunshine, Associate Professor of Anesthesiology and Pain Medicine at the UW School of Medicine, said: "It's a bit of a guttural gasping noise, and it's uniqueness makes it sound good to have a cardiac arrest."

911 calls to Seattle's emergency medical services. Because cardiac arrest patients are often unconscious, bystanders recorded the agonizing breathing 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 from the beginning of each one. 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.

"Justin Chan, Ph.D. student at the Allen School." "We also added different interfering sounds such as the sounds of cats and dogs, cars honking, air conditioning, things that you could normally hear in a home."

For the negative data, the team used 83 hours of audio data collected during sleep studies, yielding 7,305 sound samples. These clips are contained in their sleep, such as snoring or obstructive sleep apnea.

From these datasets, the team used machine learning to create a tool that could detect agonal breathing 97% of the time when the speaker was generating the sounds.

Next the team tested the algorithm to make sure it would not be accidentally classify a different type of breathing, like snoring, as agonal breathing.

"Chan said," We do not want to be alerted or urgently needed, it is important that we reduce our false positive rate, "Chan said.

For the sleep lab data, the algorithm incorrectly categorized a breathing sound as agonal breathing 0.14% of the time. The false positive was about 0.22% for separate audio clips. But when the team had the tool classify something as if it was at least 10 seconds apart, the false positive rate fell to 0% for both tests.

The team envisions this algorithm could be used as an app or a skill for Alexa that runs passively on a smart speaker or smartphone while people sleep.

"Gollakota said," This is the time to go anywhere in the world, so it's running in real time, so you do not need to do anything or anything to the cloud, "Gollakota said.

"Right now, this is a good proof of concept using the 911 calls in the Seattle metropolitan area," he said. "But we need to get access to more 911 calls related to cardiac arrest so that we can improve the accuracy of the algorithm further and ensure that it generalizes across a larger population."

The researchers plan to market this technology through UW spinout, Sound Life Sciences, Inc.

"Cardiac arrests are a very common way for people, and many of them can go unwitnessed," Sunshine said. "Part of what makes this technology so much easier?"