And so the emerging value of smartphones as a tool for diagnosing various medical conditions continues to grow. Recent advances have raised the possibility of using phones to detect ailments like ear infections, cervical cancer, HIV and syphilis. Now, researchers at the University of Washington (UW) have created an app they claim can detect sleep apnea with similar accuracy to available methods, potentially removing the need for expensive equipment and overnight hospital stays.
At best a cause of discomfort and at worst a life-threatening condition, the researchers say sleep apnea affects more than 25 million Americans, with many unaware they even have the condition. It is defined by blockages in the throat passage during sleep that can make breathing difficult. If left untreated, it can lead to something more serious, such as heart trouble, a stroke and diabetes.
Current diagnosis methods generally involve a night in hospital so the person can have their breathing monitored, along with other factors like brain activity, eye movements, heart rate and blood pressure. This sleep study is known as a polysomnogram and involves slapping a bunch of sensors and wires onto the patient's head and torso. It also costs thousands of dollars.
So how can something that fits in your pocket displace all that expensive and complex medical gear? Dubbed ApneaApp, the application repurposes an Android smartphone as a sonar system to pick up changes in breathing through abdominal, leg and other body movements. Just as bats find their way around in the dark by emitting ultrasonic sound waves and listening out for the returning echoes, the app monitors the sounds that bounce off the sleeping body.
The sound waves are emitted at a frequency that adults are unable to hear and the app filters out other noises, such as traffic or people talking. Configuring the phone for this purpose required the UW researchers to build new algorithms and develop new signal processing capabilities.
"Right now phones have sensing capabilities that we don’t fully appreciate," says Shyam Gollakota, director of the UW Networks and Mobile Systems lab. "If you can recalibrate the sensors that most phones already have, you can use them to achieve really amazing things."
The researchers put ApneaApp to the test in a clinical study involving 37 patients. They placed a Samsung Galaxy S4 on the corner of their beds across almost 300 hours of testing. This saw the app pick up on respiratory events throughout the night with 95 to 99 percent the accuracy of a polysomnogram. It also graded 32 out of the 37 subjects correctly as to the severity of their sleep apnea, as classified by what is known as the apnea-hypopnea index.
The team says that the app was shown to work properly up to 3 ft (0.91 m) away from the subject, regardless of their sleeping position or if they were under a blanket.
From here, the researchers will conduct further lab and in-home testing and look into gaining Food and Drug Administration approval for the app. The team will also investigate whether the technology could be used to pick up other body movements during sleep. It says the app could be available within the next two years.
The findings of the study will be presented at the MobiSys 2015 Conference in May in Florence, Italy and at the 2015 Associated Professional Sleep Societies meeting in Seattle in June.
You can hear from some members of the research team in the video below.
Source: University of Washington