Identify obstructive sleep apnoea using only a smartphone
ResApp has developed an obstructive sleep apnoea screening test that uses overnight breathing and snoring sounds recorded on a smartphone placed on the bedside table.
Sleep apnoea is the most common sleep breathing disorder affecting more than three in every ten men and nearly two in every ten women. Sleep apnoea is when air stops flowing to your lungs for ten seconds or longer during sleep. In some cases, this can happen more than 30 times per hour.
Studies have shown that 80% of people with sleep apnoea are undiagnosed. Sleep apnoea is today diagnosed by an overnight sleep study, often in a sleep laboratory using polysomnography (PSG). During PSG, a large number of contact sensors are used to monitor heart, lung and brain activity, breathing patterns, arm and leg movements, and blood oxygen levels whilst asleep. Patients find this uncomfortable and have difficulty sleeping in this unfamiliar environment.
Our solution requires only a smartphone placed on the bedside table. Our machine-learning algorithms analyse overnight breathing and snoring sounds to identify obstructive sleep apnoea (OSA). In a 308 patient clinical study, our algorithms correctly identified patients with OSA across the three AHI thresholds, with a sensitivity of 85%, 83% and 83% for AHI thresholds of 5/h, 15/h and 30/h respectively. At these thresholds the algorithms had a specificity of 73%, 80% and 90%. Performance was compared to a simultaneous at-home comprehensive sleep study.
Scientific Presentations & Publications
Partridge N, May J, Peltonen V, Wood, J, Keating T, Abeyratne U, Currie P, Lenstra S, Ling I, Large sample feasibility study showing smartphone‐based screening of sleep apnoea is accurate compared with polysomnography, Sleep DownUnder 2018, the 30th ASM of the Australasian Sleep Association and the Australasian Sleep Technologists Association, 2018.