Methods Machine Tools Moves Chicago Tech Center

0

Smart speakers, such as the Amazon Echo and Google Home, have proven to be adept at monitoring certain home health issues. For example, researchers from Washington University have shown that these devices can detect cardiac arrests or monitor babies’ breathing.

But what about tracking something even smaller: the minute movement of the individual heartbeats of a person sitting in front of a smart speaker?

UW researchers have developed a new skill for a smart speaker that monitors regular and irregular heartbeats without physical contact. The system sends inaudible sounds from the speaker into a room, and depending on how sounds are reflected back to the speaker, it can identify and monitor individual heartbeats. Because the heartbeat is such a tiny movement on the surface of the chest, the team’s system uses machine learning to help the smart speaker locate signals for regular and irregular heartbeats.

When the researchers tested this system on healthy participants and hospitalized heart patients, the smart speaker detected heartbeats that closely matched the beats detected by standard heart monitors. The team published these results in Communications biology.

© Mark Stone / University of Washington | https://www.washington.edu

Co-author Dr. Dan Nguyen, clinical instructor at UW School of Medicine, uses the team’s prototype smart speaker (white box in lower left corner) to demonstrate how the system works.

“Regular heartbeats are fairly easy to detect even if the signal is weak because you can look for a periodic pattern in the data,” says Shyam Gollakota, co-lead author, UW associate professor at the Paul G. Allen School of Computer Science. & Engineering. “But irregular heartbeats are really difficult because there is no such pattern. I wasn’t sure it would be possible to detect them, so I was pleasantly surprised that our algorithms could identify heartbeats. irregularities in tests with heart patients. “

While many people are familiar with the concept of heart rate, doctors are more interested in measuring heart rate. Heart rate is the average of the heartbeat over time, while a heart rate describes the pattern of the heartbeat.

For example, if a person has a heart rate of 60 beats per minute, they might have a regular heartbeat – one beat per second, or an irregular heartbeat – the beats are scattered randomly over that minute but averaging at 60 beats. per minute .

“Heart rhythm disturbances are actually more common than some of the other well-known heart conditions. Cardiac arrhythmias can cause major morbidities such as stroke, but can be very unpredictable and therefore difficult to diagnose, ”says lead co-author Dr Arun. Sridhar, assistant professor of cardiology at UW School of Medicine. “The availability of a low-cost test that can be performed frequently and at the convenience of the home can be a game-changer for some patients in terms of early diagnosis and management.”

© Mark Stone / University of Washington | https://www.washington.edu

The team’s prototype smart speaker, shown here, compares the signals from its multiple microphones (visible through the holes in this box) to identify the elusive heartbeat signal.

The key to assessing heart rate lies in identifying individual heartbeats. For this system, the heartbeat search begins when a person sits within 1 to 2 feet in front of the smart speaker. The system then emits a continuous inaudible sound, which bounces off the person and returns to the speaker. Depending on how the returned sound has changed, the system can isolate the person’s movements, including the rise and fall of their chest as they breathe.

“The movement of someone’s breath is several orders of magnitude larger on the chest wall than the movement of the heartbeat, which is quite a challenge,” says lead author Anran Wang, doctoral student at the Allen school. “And the respiratory signal is not smooth, so it’s hard to just filter it. Using the fact that smart speakers have multiple microphones, we have designed a new beamforming algorithm to help the speakers to find the heartbeat. “

The team has designed what is called a self-supervised machine learning algorithm, which learns on the fly rather than from a training package. This algorithm combines the signals from the multiple microphones of the smart speaker to identify the elusive heartbeat signal.

“It’s similar to how Alexa can still find my voice even if I’m watching a video or there are multiple people talking in the room,” says Gollakota. “When I say ‘Hey, Alexa’, the microphones are working together to find me in the room and listen to what I’m saying next. That’s basically what’s going on here, but with the heartbeat.”

The heart rate signals detected by the smart speaker do not resemble the typical peaks typically associated with traditional heart rate monitors. The researchers used a second algorithm to segment the signal into individual heartbeats so the system could extract what’s called the interval between beats, or the time between two heartbeats.

“With this method, we don’t get the electrical signal from the contracting heart. Instead, we see the vibrations on the skin when the heart beats,” says Wang.

The researchers tested a prototype smart speaker running this system on two groups: 26 healthy participants and 24 hospital patients with various heart conditions, including atrial fibrillation and heart failure. The team compared the interval between the beats of the smart speaker with that of a standard heartbeat monitor. Of the nearly 12,300 heartbeats measured for healthy participants, the median interval between smart speaker beats was within 28 milliseconds of the standard monitor. The smart speaker worked almost as well with heart patients: out of more than 5,600 heartbeats measured, the median interval between beats was less than 30 milliseconds compared to the norm.

Currently, this system is set up for spot checks: if a person is concerned about their heartbeat, they can sit in front of a smart speaker to get a reading. But the research team is hoping that future versions could continuously monitor heartbeats while people are sleeping, which could help doctors diagnose conditions like sleep apnea.

“If you have a device like this, you can monitor a patient on an extended basis and define individualized patterns for the patient. For example, we can determine when arrhythmias occur for each specific patient, and then develop corresponding care plans that are tailored when patients actually need them, ”says Sridhar. “This is the future of cardiology. And the beauty of using these types of devices is that they are already in people’s homes.”

Dr. Dan Nguyen, clinical instructor at the UW School of Medicine, is co-author of the article. This research was funded by the National Science Foundation.


Source link

Share.

Comments are closed.