As it turns out, smart speakers have proven to be handy in hospitals, allowing patients to independently control devices, as well as lessen the workload for clinicians and caretakers. And now, researchers from the University of Washington have developed an artificial intelligence system that enables these devices to monitor regular and irregular heartbeats.
Using technology to remotely and non-invasively monitor heart rate and heart rhythms isn’t new. These days most smartwatches and fitness trackers are capable of it. The neat thing here is that researchers have figured out a way to utilize the microphones in smart speakers to do it without requiring physical contact.
In a study published in Communications Biology, the researchers note that smart speakers—including Amazon Echo, Google Nest, and Apple’s HomePod—are capable of detecting where a person is in a room at a given time. In this case, the researchers had the smart speakers emit inaudible signals in the 18-22kHz range that were then reflected off a person’s body. The researchers then developed an algorithm to analyze these signals to identify subtle chest wall motions related to heartbeat, as well as separate those signals from ambient noise and breathing. Impressively, the algorithm isn’t based on a training set—it actually learns in real-time.
Put simply, a person could just sit 1-2 feet in front of a smart speaker, the speaker would play a continuous inaudible sound. That sound bounces off the person and based on how the returned sound captured by the microphone array changes, the algorithm can then figure out if you have a regular or irregular heart rhythm.
“This is similar to how Alexa can always find my voice even if I’m playing a video or if there are multiple people talking in a room,” said Shyam Gollakota, a co-senior author on the study and associate professor at the Paul G. Allen School of Computer Science & Engineering, in a statement. “When I say, ‘Hey Alexa,’ the microphones are working together to find me in the room and listen to what I say next. That’s basically what’s happening here but with the heartbeat.”
For this particular proof-of-concept setup, the researchers tested a prototype setup on 26 healthy participants and 24 hospitalized patients with various heart conditions, including atrial fibrillation and heart failure. In both cases, the smart speaker was within 28-30 milliseconds of an ECG, the gold standard used in hospitals to detect heart rate and arrhythmia.
Medically speaking, this has huge potential. The fact that it can monitor heart rhythms, without ever having to make contact with a person’s body, is big. For clarification, heart rate and heart rhythms aren’t the same things. You’re probably more familiar with the former, which is a frequently used metric in smartwatches and fitness trackers. Heart rate is the average number of heartbeats over a given time period, while heart rhythm is the pattern of those heartbeats. So you could have a perfectly normal heart rate, but irregular heart rhythms.
Like smartwatches with advanced heart features, using smart speakers in this way opens up the possibility for passive, remote heart monitoring. ECGs, while highly accurate, require a visit to the doctor and several electrodes to be placed on the body. That in turn, creates more work for hospital workers to clean and disinfect devices after each use. They’re also not capable of continuous monitoring so you’re limited to what it picks up at that exact moment in time—one reason why heart arrhythmia can be so hard to detect. Smartwatches like the Apple Watch, Samsung Galaxy Watch 3, and Fitbit Sense are capable of passive, remote, continuous monitoring, but they require you to wear the device at all times to be effective. It’s not something that’s comfortable for everyone, especially when it comes to sleep and for those with highly sensitive skin. Another issue is that these advanced smartwatches are expensive, whereas an Echo Dot, for example, is dirt cheap.
For the moment, the researchers’ system is set up for spot checks only. However, in a press release, the team notes that future versions may potentially monitor your heartbeat during sleep. That could potentially help doctors more easily identify sleep apnea, another hard-to-diagnose condition as it often requires a person to be monitored overnight at a sleep center.
“If you have a device like this, you can monitor a patient on an extended basis and define patterns that are individualized for the patient. For example, we can figure out when arrhythmias are happening for each specific patient and then develop corresponding care plans that are tailored for when the patients actually need them,” Dr. Arun Sridhar, co-senior author on the study and assistant professor of cardiology at the University of Washington School of Medicine, said. “This is the future of cardiology. And the beauty of using these kinds of devices is that they are already in people’s homes.”
As cool as this is, it’s also freaky deeky from a privacy standpoint. It’s bad enough that smart speakers are always listening in your homes. Allowing big corporations like Amazon and Google to collect your sensitive health data on top of that is a trade-off many consumers might not be keen on. On that front, the researchers note that the short-range of the frequencies used might be beneficial, as they don’t contain much information about environmental sounds. That, and the fact that third-party app developers don’t have access to “raw acoustic signals from individual microphones.” But a lot of that would be dependent on Big Tech to be proactive about data privacy in this instance—and Big Tech’s track record on smart speaker privacy ain’t great.
In any case, it’s not as if this tech will be rolled out to Amazon Echos, Google Nests, and HomePods anytime soon. As of right now, it’s merely a proof-of-concept and would require some kind of investment from the private sector to implement before becoming a reality. Still, it’s promising tech and hopefully, everyone gets their acts together so it doesn’t become a privacy nightmare.