These days, a tempting advertisement of Alexa is trending on the internet, tv and other platforms. Indeed, Alexa is a great and purposeful device that is depicting how our interaction has been changed with the machines.
But the evolution of this machine is remarkable, now this machine can help people in avoiding cardiac arrest. Yes ! this machine is potential to do this. How? Let’s take a deep look into it and see how it is emerging from the curtains.
Alexa – Amazon’s Echo Smart Speaker
Alexa is an outstanding device speaker that is based on NLP- Natural Language Processing mechanism and speech recognition. This is a machine that can understand the human command and can produce response.
It can run out searches on internet and can work as per the need of users. Now, the device has got new functionality that will make it more useful. It can help in detecting conditions of cardiac arrest. Curious to know more about it? Take a glimpse of it.
Alexa Can Suspect Cardiac Arrest
Cardiac arrests are major problem that creeps into our lives silently, and you can be unlucky enough to not sense the arrival of this situation. Thousands of people experience this situation every year, and only a few survive. Today, this is the leading cause of natural deaths. Though, the percentage of survival can surge by offering quick help.
To aid people in this situation, the company has inbuilt a tool, that uses the microphone of your smart speaker or smartphone for detecting warning signs. Altogether, it calls help on behalf of the user. Indeed, this will help in boosting survival rates.
Actually, this system has developed by expert researchers at the University of Washington. The system uses machine learning tactics and identifies the telltale gasping sound while the person is sleeping. While these sounds are known as agonal breathing, it is the situation when people struggle for air while sleeping. This gasping sound and air struggle issue are the early warning signs for a majority of cardiac arrests.
The whole mechanism to create this system is amazing. The researchers have given immense efforts in training this system, which includes the use of clips of agonal breathing. That has captured from 911 calls made in King County, Washington. They majorly used 729 calls that took 82 hours’ of recordings. Then researchers trained it on other sounds like snoring, noises people make while dealing with sleep apnea. They ensured to weed out all the false positives.
Altogether, they employed two distinct sets of recordings. The first set includes sleep sounds, that has collected by 35 volunteers, and from 12 patient – were part of sleep study because they struggled snoring and apnea. The latter recordings also produced some agonal breathing sounds to refine the accuracy of the tool.
The lead researcher of the system, Justin Chan said the when they were testing this system they found 0.2% false positive rate in the group of people who were volunteering and 0.1% rate in sleep study. This system managed to recognize the correct breathing in 97% instances.
This tool is still at the stage of proof-of-concept, so it will take many years to come in the market and be available for the public. Though, the researchers are planning to commercialize it. Chan said that there is a lot of things to do before they use it on the commercialize scale,
He also said if this were used in real life, it is potential for the system to emit 15-30 second warning users before it calls to emergency services. So, users have the chances to cancel the call in the wake of false alarm.
This device is for using in the bedroom, as most of the cardiac arrest happens in the home.
This is not enough, as there are numbers of challenges that need to be overcome before the final launch of the device. The improvements include privacy as a chief concern.
Indeed, this device will bring a huge shift in the cases of heart arrest. To know more, stay tuned with Techugo.2