The World Health Organization describes obstructive sleep apnea syndrome as "a clinical disorder marked by frequent pauses in breathing during sleep usually accompanied by loud snoring." Close to one billion people around the world suffer from it, unable to sleep without waking up multiple times every night.
Now a prototype chest patch by Onera Health relies on machine learning and bioimpedence measurements to detect sleep apnea.
According to a July 2019 medical report authored by several physicians and doctors from around the world, an estimated 970 million people suffer from obstructive sleep apnea around the world.
As mentioned above, sleep apnea is caused by uncontrolled pausing in breathing. These pauses cut off the oxygen supply to your body for a few seconds and halt the removal of carbon dioxide. As a result of this, your brain briefly wakes you up, re-opens the airways and re-starts breathing, according to the World Health Organization.
Getting tested for sleep apnea is a cumbersome process right now, usually results in spending a night in a clinic connected to sensors that measure your brain activity, eye movement, and blood oxygen levels. But Onera Health's device patch hopes to make all this history.
The device appears to be tinier than a credit card, described in a study published 20?January in the IEEE Journal of Biomedical and Health Informatics,which measures electrical signals passing through the?body (known as bioimpedance) and machine learning algorithms to effectively predict sleep apnea attacks. It's also supposed to be a convenient monitoring of sleep apnea, with a heavy impetus on tracking your breathing.
Several sleep apnea detecting devices are already there in the market, but they aren't very effective. This led researchers a group of researchers felt they could do much better, and their device, called Robin, does so by applying a small current to the body at a known frequency, and measures the resulting voltage at a different location after it passes through the body, according to an IEEE Spectrum report.
The researchers quickly realized that Robin can be used to fairly accurately monitor a wearer’s breathing. The team used AI and machine learning to the bioimpedance measurements to detect sleep apnea events. Their approach has an accuracy of 73?percent at detecting sleep apnea events among 25 volunteers, which is a pretty good success rate.
The team behind Robin believes their respiratory technique based sleep apnea monitoring can be used to detect and monitor other breathing-based disorders as well.