Use of smartphone's microphone sensor to estimate rate of respiration

Abstract

Sleep plays a crucial part of one’s physical and mental wellbeing. Polysomnography, which is the gold standard for measuring and analyzing sleep requires the subject to sleep overnight with many expensive sensors attached to the body. A disordered respiration pattern can lead to different physiological and psychological dysfunctioning, in turn hampering the sleep. PSG uses specialized flow or pressure sensor for measuring rate of respiration. This thesis proposes a portable, comfortable, and cost-effective solution using the smartphone to estimate respiration rate with the help of attached headset microphone placed beneath the nose. The method uses the amplitude of nasal breath sound to estimate different breathing rates. Inhale and exhale periods are estimated by taking the average of the amplitude of nasal breath sound. Performance evaluation of the system involved collecting data from people of different gender and across all age groups. The experimental results showed that the method achieved good accuracy with a median error of 10% for varied breathing rates