Monitoring sound to quantify snoring and sleep apnea severity using a smartphone: proof of concept

Abstract

Study objectives: Habitual snoring is a prevalent condition that is not only a marker of obstructive sleep apnea (OSA) but can also lead to vascular risk. However, it is not easy to check snoring status at home. We attempted to develop a snoring sound monitor consisting of a smartphone alone, which is aimed to quantify snoring and OSA severity.

Methods: The subjects included 50 patients who underwent diagnostic polysomnography (PSG), of which the data of 10 patients were used for developing the program and that of 40 patients were used for validating the program. A smartphone was attached to the anterior chest wall over the sternum. It acquired ambient sound from the built-in microphone and analyzed it using a fast Fourier transform on a real-time basis.

Results: Snoring time measured by the smartphone highly correlated with snoring time measured by PSG (r = 0.93). The top 1 percentile value of sound pressure level (L1) determined by the smartphone correlated with the ambient sound L1 during sleep determined by PSG (r = 0.92). Moreover, the respiratory disturbance index estimated by the smartphone (smart-RDI) highly correlated with the apnea-hypopnea index (AHI) obtained by PSG (r = 0.94). The diagnostic sensitivity and specificity of the smart-RDI for diagnosing OSA (AHI ≥ 15) were 0.70 and 0.94, respectively.

Conclusions: A smartphone can be used for effectively monitoring snoring and OSA in a controlled laboratory setting. Use of this technology in a noisy home environment remains unproven, and further investigation is needed.

Keywords: Snoring; home monitoring; obstructive sleep apnea; smartphone.

Figure 1. Representative trace showing relationship between polysomnograph signals and smartphone-acquired sounds

(A) Polysomnography signals. One obstructive apnea (OA) and 3 obstructive hypopnea (OH) events are detected during the 5 minutes data. Tracheal sound spectrogram shows snoring (indicated by red color) predominantly during hypopnea events. (B) Smartphone-acquired sounds. The parameters for detection of snoring (total power, peak power spectral density) and respiratory events (filtered total power time series) are shown. Arrows indicate the sound power dips detected at a threshold of 3 dB (see text).

Figure 2. The relationship between the equivalent sound pressure level (Leq) determined by the smartphone and that of tracheal sounds and ambient sounds during sleep determined by polysomnography

Figure 3. The relationship between the top 1 percentile sound pressure level (L1) determined by the smartphone and that of tracheal sounds and ambient sounds during sleep determined by polysomnography

Figure 4. The relationship between the snoring time determined by the smartphone and that determined by polysomnography (tracheal sounds)

Figure 5. The relationship between the respiratory disturbance index determined by the smartphone (smart-RDI) and the apnea-hypopnea index (AHI) determined by polysomnography

Figure 6. Bland-Altman plots showing variance between the respiratory disturbance index determined by the smartphone (smart-RDI) and the apnea-hypopnea index (AHI) determined by polysomnography

Black horizontal lines indicate 95% limits of agreement.

Figure 7. Receiver operating characteristic curve showing the relationship between diagnostic sensitivity and specificity of the respiratory disturbance index by smartphone (smart-RDI) in the validation group when the following cutoff values are used: (A) apnea-hypopnea index (AHI) ≥ 15, (B) AHI ≥ 30