Cardiac arrest is one of the major causes of natural deaths in the US and almost 400,000 suffer one cardiac arrest every year in the country. Death rates from cardiac arrest can be minimized by timely detection and immediate resuscitation. Keeping this mind, a team of researchers from the University of Washington has developed a new tool that could be fitted into a smart speaker to detect early signs of cardiac arrest by listening to your breathing.
The tool deploys machine learning and spots telltale gasping sound or agonal breathing from the sound of your breathing. Agonal sound is the sound that people make when they are struggling to breathe. In more than half of the cases, it could be an early warning for a cardiac arrest.
Researchers have trained the tool by making it listen to agonal breathing sounds from 911 calls made by the residents of King County, Washington. The model has been trained on 729 calls amounting to 82 hours of recordings.
To make the system foolproof, they also trained the tool on other surrounding sounds including the sound of snoring to cancel out the false positives.
The results are very promising as the system was able to distinguish agonal breathing in 97% of the cases from up to 20 feet away.
The tool once fitted in a smart speaker could call an emergency service on your behalf in case of a possible cardiac arrest detection.
However, the tool is far from being commercially available as it is still at the proof-of-concept stage and it will take years until we see the tool in a smart speaker around us.
There are some major challenges involved, like privacy as the tool will hear your sound all the time to detect a cardiac arrest.
According to Peter Chai, an assistant professor of emergency medicine at Brigham and Women’s Hospital in Boston, “There are questions around what you do with ambient noise of others in a room, or if you’re gathering information from a phone’s microphone, or what you do with inadvertent recording.”
Nonetheless, it could be a great tool to save the lives of cardiac arrest sufferers.