Spectral analysis of snoring events from an Emfit mattress

Abstract

The aim of this study is to explore the capability of an Emfit (electromechanical film transducer) mattress to detect snoring (SN) by analyzing the spectral differences between normal breathing (NB) and SN. Episodes of representative NB and SN of a maximum of 10 min were visually selected for analysis from 33 subjects. To define the bands of interest, we studied the statistical differences in the power spectral density (PSD) between both breathing types. Three bands were selected for further analysis: 6-16 Hz (BW1), 16-30 Hz (BW2) and 60-100 Hz (BW3). We characterized the differences between NB and SN periods in these bands using a set of spectral features estimated from the PSD. We found that 15 out of the 29 features reached statistical significance with the Mann-Whitney U-test. Diagnostic properties for each feature were assessed using receiver operating characteristic analysis. According to our results, the highest diagnostic performance was achieved using the power ratio between BW2 and BW3 (0.85 area under the receiver operating curve, 80% sensitivity, 80% specificity and 80% accuracy). We found that there are significant differences in the defined bands between the NB and SN periods. A peak was found in BW3 for SN epochs, which was best detected using power ratios. Our work suggests that it is possible to detect snoring with an Emfit mattress. The mattress-type movement sensors are inexpensive and unobtrusive, and thus provide an interesting tool for sleep research.

Detection of Snores Using Source Separation on an Emfit Signal

“…Due to our initialization procedure of the NMFD algorithm, snoring always corresponds to the second component, 2 [ ]. This spectral profile is in line with our previous study [32] where a similar power spectrum density (PSD) was found when comparing snoring and normal breathing PSDs. It is also consistent why power ratios perform well in certain SN profiles [44].…”

“…4. In our previous study [32], we found statistical differences in the bands 6-16 Hz and 60-100 Hz. These spectral components in 6-16 Hz are very likely to be related to increased efforts, which often appear concurrently with SN.…”

“…In this study, we initialized each atom of matrix A to those frequencies that matches our source signal. Previous work [32], found a significant power increase in the bands and Hz, thus we initialized the 2 components to be maximum at frequencies at 30-100 Hz, and 1 components to be maximum at 6-30 Hz:…”

“…In our previous study, we characterized the spectral differences of thirty-seconds epochs length from NB and SN periods in the Emfit signal using the power spectrum [32]. We wanted to continue to develop unsupervised methods to detect individual snores based on source separation techniques using spectrogram factorization.…”

The Supportive Smart Home System: Implications and Solutions for Service Providers

“…• EMFIT QS: It is installed to assess bed occupancy, time in bed, vital signs, and sleep state [63], [65] and uses a Wi-Fi connection to access the Internet. The resulting network configuration is shown in Figure 4, where the deployment focused on a method to allow the traffic from all of the systems and sensors under test to be monitored and measured while not burdening the system with the measurement of any traffic from other devices deployed within the residence for personal use, school-at-home, and work-at-home during the pandemic.…”