Medical

Wrist-worn device used to more easily assess sleep apnea

Wrist-worn device used to more...
A test subject wears both the device and traditional sensors while sleeping at the Kempenhaeghe Epilepsy and Sleep Center
A test subject wears both the device and traditional sensors while sleeping at the Kempenhaeghe Epilepsy and Sleep Center
View 1 Image
A test subject wears both the device and traditional sensors while sleeping at the Kempenhaeghe Epilepsy and Sleep Center
1/1
A test subject wears both the device and traditional sensors while sleeping at the Kempenhaeghe Epilepsy and Sleep Center

The assessment of sleep apnea typically involves spending a night at a clinic, wired up to various sensors that may actually keep the patient from sleeping normally. According to recent research, though, a Fitbit-like device could serve the same purpose while the patient sleeps at home.

The study was led by Gabriele Papini, a PhD researcher at the Netherlands' Eindhoven University of Technology.

It incorporated a wrist-worn device, similar to a fitness tracker, which shines green LED light through the user's skin and into the underlying blood vessels. By analyzing how much of that light is absorbed by the blood and how much is reflected back up to the underside of the device, it's possible to continuously measure the wearer's heart rate in real time.

Papini and colleagues believed that changes in heart rate could correspond to changes in respiration caused by sleep apnea. If so, then patients could comfortably wear the device for multiple nights while sleeping in their own bed, providing more and better data than if they just spent one night at a sleep clinic, wired up to multiple sensors.

The scientists started by using the device to monitor the heart rate and pulse amplitude of 250 volunteers, some of whom were known to suffer from sleep apnea, and some of whom were known not to. That data was then used to train deep-learning-based algorithms.

Those algorithms were subsequently able to match tell-tale changes in heart rate/pulse amplitude to apnea-induced respiratory incidents, plus they also learned to filter out distracting "background noise" such as body movements. As a result, it was possible to calculate what is known as an "apnea-hypopnea index" – which is the number of unusual respiratory events per hour of sleep – for each person.

When the device and the algorithms were tested on another 250 volunteers, the calculated index for each person was found to fall closely in line with one obtained utilizing traditional sensors of the type commonly used in sleep clinics.

"Hopefully, this research will lead to new techniques that, in addition to a better diagnosis, can also check on the efficiency of treatments for patients with sleep disorders," said Papini's main supervisor, Prof. Sebastiaan Overeem. "And importantly, the device could be used at home and for prolonged periods of time."

Source: Eindhoven University of Technology

No comments
0 comments
There are no comments. Be the first!