Continuous analysis and monitoring of snores and their relationship to the apnea-hypopnea index

Obstructive Sleep Apnea Detection Based on Sleep Sounds via Deep Learning

PURPOSE

This study aimed to propose a novel deep-learning method for automatic sleep apneic event detection and thus to estimate the apnea hypopnea index (AHI) and identify obstructive sleep apnea (OSA) in an event-by-event manner solely based on sleep sounds obtained by a noncontact audio recorder.

METHODS

We conducted a cross-sectional study of participants with habitual snoring or heavy breathing sounds during sleep to train and test a deep convolutional neural network named OSAnet for the detection of OSA based on sleep sounds. Polysomnography (PSG) was conducted, and sleep sounds were recorded simultaneously in a regular room without noise attenuation. The study was conducted in two phases. In phase one, eligible participants were enrolled and randomly allocated into training and validation groups for deep learning algorithm development. In phase two, eligible patients were enrolled in a test group for algorithm assessment. Sensitivity, specificity, accuracy, unweighted Cohen kappa coefficient (κ) and the area under the curve (AUC) were calculated using PSG as the reference standard.

RESULTS

A total of 135 participants were randomly divided into a training group (n, 116) and a validation group (n, 19). An independent test group of 59 participants was subsequently enrolled. Our algorithm achieved a precision of 0.81 and sensitivity of 0.78 in the test group for overall sleep event detection. The algorithm exhibited robust diagnostic performance to identify severe cases with a sensitivity of 95.6% and specificity of 91.6%.

CONCLUSION

Our results showed that a deep learning algorithm based on sleep sounds recorded by a noncontact voice recorder served as a feasible tool for apneic event detection and OSA identification. This technique may hold promise for OSA assessment in the community in a relatively comfortable and low-cost manner. Further studies to develop a tool based on a home-based setting are warranted.

Relationship between Sleep Stages and HRV response in Obstructive Sleep Apnea Patients

Patients suffering from obstructive sleep apnea (OSA) usually present an increased sympathetic activity caused by the intermittent hypoxia effect on autonomic control. This study evaluated the relationship between sleep stages and the apnea duration, frequency, and type, as well as their impact on HRV markers in different groups of disease severity. The hypnogram and R-R interval signals were extracted in 81 OSA patients from night polysomnographic (PSG) recordings. The apnea-hypopnea index (AHI) defined patient classification as mild-moderate (AHI<=30, n=44) or severe (AHI>30, n=37). The normalized power in VLH, LF, and HF bands of RR series were estimated by a time-frequency approach and averaged in 1-min epochs of normal and apnea segments. The autonomic response and the impact of sleep stages were assessed in both segments to compare patient groups. Deeper sleep stages (particularly S2) concentrated the shorter and mild apnea episodes (from 10 to 40 s) compared to light (SWS) and REM sleep. Longer episodes (>50 s) although less frequent, were of similar incidence in all stages. This pattern was more pronounced for the group of severe patients. Moreover, during apnea segments, LF_nu was higher (p=0.044) for the severe group, since V LF_nu and HF_nu presented the greatest changes when compared to normal segments. The non-REM sleep seems to better differentiate OSA patients groups, particularly through VLF_nu and HF_nu(p<0.001). A significant difference in both sympathetic and vagal modulation between REM and non-REM sleep was only found within the severe group. These results confirm the importance of considering sleep stages for HRV analysis to further assess OSA disease severity, beyond the traditional and clinically limited AHI values.Clinical relevance-Accounting for sleep stages during HRV analysis could better assess disease severity in OSA patients.

A Review of Methods for Sleep Arousal Detection Using Polysomnographic Signals

Multiple types of sleep arousal account for a large proportion of the causes of sleep disorders. The detection of sleep arousals is very important for diagnosing sleep disorders and reducing the risk of further complications including heart disease and cognitive impairment. Sleep arousal scoring is manually completed by sleep experts by checking the recordings of several periods of sleep polysomnography (PSG), which is a time-consuming and tedious work. Therefore, the development of efficient, fast, and reliable automatic sleep arousal detection system from PSG may provide powerful help for clinicians. This paper reviews the automatic arousal detection methods in recent years, which are based on statistical rules and deep learning methods. For statistical detection methods, three important processes are typically involved, including preprocessing, feature extraction and classifier selection. For deep learning methods, different models are discussed by now, including convolution neural network (CNN), recurrent neural network (RNN), long-term and short-term memory neural network (LSTM), residual neural network (ResNet), and the combinations of these neural networks. The prediction results of these neural network models are close to the judgments of human experts, and these methods have shown robust generalization capabilities on different data sets. Therefore, we conclude that the deep neural network will be the main research method of automatic arousal detection in the future.

Prediction of Apnea-Hypopnea Index Using Sound Data Collected by a Noncontact Device

OBJECTIVE

To predict the apnea-hypopnea index (AHI) in patients with obstructive sleep apnea (OSA) using data from breathing sounds recorded using a noncontact device during sleep.

STUDY DESIGN

Prospective cohort study.

SETTING

Tertiary referral hospital.

SUBJECT AND METHODS

Audio recordings during sleep were performed using an air-conduction microphone during polysomnography. Breathing sounds recorded from all sleep stages were analyzed. After noise reduction preprocessing, the audio data were segmented into 5-second windows and sound features were extracted. Estimation of AHI by regression analysis was performed using a Gaussian process, support vector machine, random forest, and simple linear regression, along with 10-fold cross-validation.

RESULTS

In total, 116 patients who underwent attended, in-laboratory, full-night polysomnography were included. Overall, random forest resulted in the highest performance with the highest correlation coefficient (0.83) and least mean absolute error (9.64 events/h) and root mean squared error (13.72 events/h). Other models resulted in somewhat lower but similar performances, with correlation coefficients ranging from 0.74 to 0.79. The estimated AHI tended to be underestimated as the severity of OSA increased. Regarding bias and precision, estimation performances in the severe OSA subgroup were the lowest, regardless of the model used. Among sound features, derivative of the area methods of moments of overall standard deviation demonstrated the highest correlation with AHI.

CONCLUSION

AHI was fairly predictable by using data from breathing sounds generated during sleep. The prediction model may be useful not only for prescreening but also for follow-up after treatment in patients with OSA.

Objective Relationship Between Sleep Apnea and Frequency of Snoring Assessed by Machine Learning

STUDY OBJECTIVES

Snoring is perceived to be directly proportional to sleep apnea severity, especially obstructive sleep apnea (OSA), but this notion has not been thoroughly and objectively evaluated, despite its popularity in clinical practice. This might lead to overdiagnosis or underdiagnosis of OSA. The goal of this study is to examine this notion and objectively quantify the relationship between sleep apnea and snoring detected using advanced signal processing algorithms.

METHODS

We studied adults referred for polysomnography, from which the apnea-hypopnea index (AHI) was derived. Breath sounds were recorded simultaneously, from which snoring was accurately quantified using acoustic analysis of breath sounds and machine-learning computer algorithms. The snore index (SI) was calculated as the number of snores per hour of sleep.

RESULTS

In 235 patients, the mean AHI was 20.2 ± 18.8 and mean SI was 320.2 ± 266.7 events/h. On the one hand, the overall correlation between SI and AHI was weak but significant (r = .32, P < .0001). There was a significant stepwise increase in SI with increasing OSA severity, but with a remarkable overlap in SI among OSA severity categories. On the other hand, SI had weak negative correlation with central AHI (r = -.14, P = .035). SI had modest positive and negative predictive values for OSA (0.63 and 0.62 on average, respectively) and good sensitivity but low specificity (0.91 and 0.31 on average, respectively) attributed to the large number of snorers without OSA.

CONCLUSIONS

Snoring on its own is probably of limited usefulness in assessing sleep apnea presence and severity, because of its weak relationship with AHI. Thus, the complaint of snoring should be interpreted with caution to avoid unnecessary referrals for sleep apnea testing. Conversely, clinicians should be aware of the possibility of missing diagnosis of patients with sleep apnea who have minimal snoring.

Predictors of Sleep Apnea in the Canadian Population

Older age, obesity, hypertension, snoring, and excessive daytime sleepiness have been associated with sleep apnea. The objective of this study was to determine the prevalence (crude and adjusted), as well as the risk factors, of sleep apnea in the adult Canadian population. Data from the 2009 Sleep Apnea Rapid Response (SARR) questionnaire were used to identify the risk factors, and all sleep-related questions in the SARR questionnaire were used. The outcome variable of interest was health professional-diagnosed sleep apnea. Covariates of interest were demographic variables, population characteristics, respiratory and cardiovascular diseases, and enabling resources. The multiple logistic regression model adjusted for the clustering effect was used to analyze the data. Sleep apnea was diagnosed in 858,913 adults (3.4% of the population), and more men (65.4%) than women (34.6%) were diagnosed with sleep apnea. Multivariable logistic regression analysis indicated that age (45 and older), loud snoring, sudden awakening with gasping/choking (rare/sometimes and once or more a week), and nodding off/falling asleep in driving in the past 12 months were significantly associated with diagnosed sleep apnea. Predictive probability demonstrated that in overweight and obese persons, ≥15 minutes of daily exercise significantly decreased the risk of diagnosed sleep apnea. The conclusion of this study is that in the Canadian population, sleep apnea is associated with older age, loud snoring, and sleeping problems. The protective effect of exercise warrants further investigation.

mHealth tools for monitoring Obstructive Sleep Apnea patients at home: Proof-of-concept

Obstructive Sleep Apnea (OSA) is a sleep disorder that affects mainly the adult and elderly population. Due to the high percentage of patients who remain undiagnosed and untreated because of limitations of current diagnosis methods, the management of OSA is an important social, scientific and economic problem that will be difficult to be assumed by health systems. On the other hand, smartphone platforms (mHealth systems) are being considered as an innovative solution, thanks to the integration of the essential sensors to obtain clinically relevant parameters in the same device or in combination with wireless wearable devices.

Characterization of a tooth microphone coupled to an oral appliance device: A new system for monitoring OSA patients

Obstructive sleep apnea (OSA) is a highly prevalent chronic disease, especially in elderly and obese populations. Despite constituting a serious health, social and economic problem, most patients remain undiagnosed and untreated due to limitations in current equipment. In this work, we propose a novel method to diagnose OSA and monitor therapy adherence and effectiveness at home in a non-invasive and inexpensive way: combining acoustic analysis of breathing and snoring sounds with oral appliance therapy (OA). Audiodontics has introduced a new sensor, a tooth microphone coupled to an OA device, which is the main pillar of this system. The objective of this work is to characterize the response of this sensor, comparing it with a commercial tracheal microphone (Biopac transducer). Signals containing OSA-related sounds were acquired simultaneously with the two microphones for that purpose. They were processed and analyzed in time, frequency and time-frequency domains, in a custom MATLAB interface. We carried out a single-event approach focused on breaths, snores and apnea episodes. We found that the quality of the signals obtained by both microphones was quite similar, although the tooth microphone spectrum concentrated more energy at the high-frequency band. This opens a new field of study about high-frequency components of snores and breathing sounds. These characteristics, together with its intraoral position, wireless option and combination with customizable OAs, give the tooth microphone a great potential to reduce the impact of sleep disorders, by enabling prompt detection and continuous monitoring of patients at home.

Detection of sleep disordered breathing severity using acoustic biomarker and machine learning techniques

PURPOSE

Breathing sounds during sleep are altered and characterized by various acoustic specificities in patients with sleep disordered breathing (SDB). This study aimed to identify acoustic biomarkers indicative of the severity of SDB by analyzing the breathing sounds collected from a large number of subjects during entire overnight sleep.

METHODS

The participants were patients who presented at a sleep center with snoring or cessation of breathing during sleep. They were subjected to full-night polysomnography (PSG) during which the breathing sound was recorded using a microphone. Then, audio features were extracted and a group of features differing significantly between different SDB severity groups was selected as a potential acoustic biomarker. To assess the validity of the acoustic biomarker, classification tasks were performed using several machine learning techniques. Based on the apnea-hypopnea index of the subjects, four-group classification and binary classification were performed.

RESULTS

Using tenfold cross validation, we achieved an accuracy of 88.3% in the four-group classification and an accuracy of 92.5% in the binary classification. Experimental evaluation demonstrated that the models trained on the proposed acoustic biomarkers can be used to estimate the severity of SDB.

CONCLUSIONS

Acoustic biomarkers may be useful to accurately predict the severity of SDB based on the patient's breathing sounds during sleep, without conducting attended full-night PSG. This study implies that any device with a microphone, such as a smartphone, could be potentially utilized outside specialized facilities as a screening tool for detecting SDB.

Characterization of microphones for snoring and breathing events analysis in mHealth

Obstructive sleep apnea (OSA) is one of the most common sleep disorders, especially in elderly population. Despite its high prevalence and severe consequences, most patients remain undiagnosed due to serious limitations on the existing equipment. Efforts are being done to find cost-effective alternatives and mHealth solutions could play a key role. One promising approach in this context is the acoustic analysis of snoring. The sensor it requires is a microphone, which is widely available in different models and even integrated in smartphones. The objective of this work is to characterize and compare the responses of two commercial tracheal microphones and a mHealth-based microphone, as a proof-of-concept to evaluate their potential as sensors for OSA detection. To do that, we designed an experimental protocol to study OSA-related events (breaths, snores and apneas) simulated by 4 subjects. Test signals were simultaneously recorded with different microphones and posteriorly processed and analyzed. We accurately characterized the frequency response of the two commercial microphones, finding that one of them was too restrictive (bandwidth 50-250 Hz) and thus not suitable as snoring sensor for high-frequency acoustic analysis. Regarding smartphones, we studied the Samsung Galaxy S5 microphone. We found that, when located over the thorax, it provided quality signals comparable to those of tracheal microphones, with a broader frequency response. Further work is required, but this preliminary study suggests that acoustic analysis of snoring through mHealth solutions can be a feasible alternative to screen and monitor OSA patients at home.

Detecting Breathing and Snoring Episodes Using a Wireless Tracheal Sensor-A Feasibility Study

OBJECTIVE

Sleep-disordered breathing is both a clinical and a social problem. This implies the need for convenient solutions to simplify screening and diagnosis. The aim of the study was to investigate the sensitivity and specificity of a novel wireless system in detecting breathing and snoring episodes during sleep.

METHODS

A wireless acoustic sensor was elaborated and implemented. Segmentation (based on spectral thresholding and heuristics) and classification of all breathing episodes during recording were implemented through a mobile application. The system was evaluated on 1520 manually labeled episodes registered from 40 real-world, whole-night recordings of 16 generally healthy subjects.

RESULTS

The differentiation between normal breathing and snoring had 88.8% accuracy. As the system is intended for screening, high specificity of 95% is reported.

CONCLUSION

The system is a compromise between nonmedical phone applications and medical sleep studies. The presented approach enables the study to be repetitive, personal, and inexpensive. It has additional value in the form of well-recorded data which are reliable and comparable.

SIGNIFICANCE

The system opens unexplored possibilities in sleep monitoring and study enabling a multinight recording strategy involving the collection and analysis of abundant data from thousands of people.

Automatic Video Analysis for Obstructive Sleep Apnea Diagnosis

STUDY OBJECTIVES

We investigated the diagnostic accuracy for the identification of obstructive sleep apnea (OSA) and its severity of a noninvasive technology based on image processing (SleepWise).

METHODS

This is an observational, prospective study to evaluate the degree of agreement between polysomnography (PSG) and SleepWise. We recruited 56 consecutive subjects with suspected OSA who were referred as outpatients to the Sleep Unit of the Hospital Universitari Germans Trias i Pujol (HUGTiP) from January 2013 to January 2014. All patients underwent laboratory PSG and image processing with SleepWise simultaneously the same night. Both PSG and SleepWise analyses were carried independently and blindly.

RESULTS

We analyzed 50 of the 56 patients recruited. OSA was diagnosed through PSG in a total of 44 patients (88%) with a median apnea-hypopnea index (AHI) of 25.35 (24.9). According to SleepWise, 45 patients (90%) met the criteria for a diagnosis of OSA, with a median AHI of 22.8 (22.03). An analysis of the ability of PSG and SleepWise to classify patients by severity on the basis of their AHI shows that the two diagnostic systems distribute the different groups similarly. According to PSG, 23 patients (46%) had a diagnosis of severe OSA, 11 patients (22%) moderate OSA, and 10 patients (20%) mild OSA. According to SleepWise, 20, 13, and 12 patients (40%, 26%, and 24%, respectively) had a diagnosis of severe, moderate, and mild OSA respectively. For OSA diagnosis, SleepWise was found to have sensitivity of 100% and specificity of 83% in relation to PSG. The positive predictive value was 97% and the negative predictive value was 100%. The Bland-Altman plot comparing the mean AHI values obtained through PSG and SleepWise shows very good agreement between the two diagnostic techniques, with a bias of -3.85, a standard error of 12.18, and a confidence interval of -0.39 to -7.31.

CONCLUSIONS

SleepWise was reasonably accurate for noninvasive and automatic diagnosis of OSA in outpatients. SleepWise determined the severity of OSA with high reliability. The current study including simultaneous laboratory PSG and SleepWise processing image is proposed as a reasonable validation standard.

The Frequency and Energy of Snoring Sounds Are Associated with Common Carotid Artery Intima-Media Thickness in Obstructive Sleep Apnea Patients

Obstructive sleep apnea (OSA) is a known risk factor for atherosclerosis. We investigated the association of common carotid artery intima-media thickness (CCA-IMT) with snoring sounds in OSA patients. A total of 30 newly diagnosed OSA patients with no history of cardiovascular diseases were prospectively enrolled for measuring mean CCA-IMT with B-mode ultrasonography, body mass index, metabolic syndrome, 10-year cardiovascular disease risk score, high-sensitivity C-reactive protein, and homocysteine. Good-quality signals of full-night snoring sounds in an ordinary sleep condition obtained from 15 participants were further acoustically analyzed (Included group). All variables of interest were not significantly different (all p > 0.05) between the included and non-included groups except for diastolic blood pressure (p = 0.037). In the included group, CCA-IMT was significantly correlated with snoring sound energies of 0–20 Hz (r = 0.608, p = 0.036) and 652–1500 Hz (r = 0.632, p = 0.027) and was not significantly associated with that of 20–652 Hz (r = 0.366, p = 0.242) after adjustment for age and sex. Our findings suggest that underlying snoring sounds may cause carotid wall thickening and support the large-scale evaluation of snoring sound characters as markers of surveillance and for risk stratification at diagnosis.

Snoring Sounds Predict Obstruction Sites and Surgical Response in Patients with Obstructive Sleep Apnea Hypopnea Syndrome

Snoring sounds generated by different vibrators of the upper airway may be useful indicators of obstruction sites in patients with obstructive sleep apnea hypopnea syndrome (OSAHS). This study aimed to investigate associations between snoring sounds, obstruction sites, and surgical responses (≥50% reduction in the apnea-hypopnea index [AHI] and <10 events/hour) in patients with OSAHS. This prospective cohort study recruited 36 OSAHS patients for 6-hour snoring sound recordings during in-lab full-night polysomnography, drug-induced sleep endoscopy (DISE), and relocation pharyngoplasty. All patients received follow-up polysomnography after 6 months. Fifteen (42%) patients with at least two complete obstruction sites defined by DISE were significantly, positively associated with maximal snoring sound intensity (40-300 Hz; odds ratio [OR], 1.25, 95% confidence interval [CI] 1.05-1.49) and body mass index (OR, 1.48, 95% CI 1.02-2.15) after logistic regression analysis. Tonsil obstruction was significantly, inversely correlated with mean snoring sound intensity (301-850 Hz; OR, 0.84, 95% CI 0.74-0.96). Moreover, baseline tonsil obstruction detected by either DISE or mean snoring sound intensity (301-850 Hz), and AHI could significantly predict the surgical response. Our findings suggest that snoring sound detection may be helpful in determining obstruction sites and predict surgical responses.

Predicting Obstructive Sleep Apnea with Periodic Snoring Sound Recorded at Home

STUDY OBJECTIVES

The cost-effectiveness of diagnosing obstructive sleep apnea (OSA) could be improved by using a preliminary screening method among subjects with no suspicion of other sleep disorders. We aimed to evaluate the diagnostic value of periodic snoring sound recorded at home.

METHODS

We included 211 subjects, aged 18-83 (130 men), who were referred to our laboratory for suspicion of OSA, and had a technically successful overnight polygraphy, measured with the Nox T3 Sleep Monitor (Nox Medical, Iceland) with a built-in microphone. We analyzed the percentage of periodic snoring during the home sleep apnea study.

RESULTS

Apnea-hypopnea index (AHI) ranged from 0.1 to 116 events/h and the percentage of periodic snoring from 1% to 97%. We found a strong positive correlation (r = 0.727, p < 0.001) between periodic snoring and AHI. The correlation was slightly stronger among female, younger, and obese subjects. The best threshold value of the periodic snoring for predicting an AHI > 15 events/h with as high sensitivity as possible was found to be 15%. There, sensitivity was 93.3%, specificity 35.1%, and negative predictive value 75.0%.

CONCLUSIONS

According to our results, it is possible to set a periodic snoring threshold (15% or more) for the subject to advance to further sleep studies. Together with medical history and prior to more expensive studies, measuring periodic snoring at home is a simple and useful method for predicting the probability of OSA, in particular among women who are often unaware of their apnea-related snoring.

Antioxidant Carbocysteine Treatment in Obstructive Sleep Apnea Syndrome: A Randomized Clinical Trial

Objective

This study aimed to examine the effects of carbocysteine in OSAS patients.

Methods

A total of 40 patients with moderate to severe obstructive sleep apnea syndrome (OSAS) were randomly divided into two groups. One group was treated with 1500 mg carbocysteine daily, and the other was treated with continuous positive airway pressure (CPAP) at night. Before treatment and after 6 weeks of treatment, all patients underwent polysomnography and completed questionnaires. Treatment compliance was compared between the two groups. Plasma was collected for various biochemical analyses. Endothelial function was assessed with ultrasound in the carbocysteine group.

Results

The proportion of patients who fulfilled the criteria for good compliance was higher in the carbocysteine group (n = 17) than in the CPAP group (n = 11; 100% vs. 64.7%). Compared with baseline values, the carbocysteine group showed significant improvement in their Epworth Sleepiness Scale score (10.18±4.28 vs. 6.82±3.66; P≤0.01), apnea-hypopnea index (55.34±25.03 vs. 47.56±27.32; P≤0.01), time and percentage of 90% oxygen desaturation (12.66 (2.81; 50.01) vs. 8.9 (1.41; 39.71); P≤0.01), and lowest oxygen saturation level (65.88±14.86 vs. 70.41±14.34; P≤0.01). Similar changes were also observed in the CPAP group. The CPAP group also showed a decreased oxygen desaturation index and a significant increase in the mean oxygen saturation after treatment, but these increases were not observed in the carbocysteine group. Snoring volume parameters, such as the power spectral density, were significantly reduced in both groups after the treatments. The plasma malondialdehyde level decreased and the superoxide dismutase and nitric oxide levels increased in both groups. The endothelin-1 level decreased in the CPAP group but did not significantly change in the carbocysteine group. Ultrasonography showed that the intima-media thickness decreased (0.71±0.15 vs. 0.66±0.15; P≤0.05) but that flow-mediated dilation did not significantly change in the carbocysteine group.

Conclusions

Oral carbocysteine slightly improves sleep disorders by attenuating oxidative stress in patients with moderate to severe OSAS. Carbocysteine may have a role in the treatment of OSAS patients with poor compliance with CPAP treatment. However, the efficiency and feasibility of carbocysteine treatment for OSAS needs further evaluation.

Trial Registration

ClinicalTrials.gov NCT02015598

Relationship Between Snoring Intensity and Severity of Obstructive Sleep Apnea

Objectives

The aim of this study was to determine the relationship between the intensity of snoring and severity of sleep apnea using Watch-PAT (peripheral arterial tone) 100.

Methods

A total of 404 patients (338 males and 66 females) who underwent home-based portable sleep study using Watch-PAT 100 for obstructive sleep apnea (OSA) from January 2009 through December 2011 were included in this study. Subjects were divided into 4 groups; no OSA (PAT apnea hypopnea index [pAHI]<5/hour), mild OSA (5≤pAHI<15/hour), moderate OSA (15≤pAHI<30/hour), or severe OSA groups (pAHI≥30/hour). Mean snoring intensity and percent sleep time with snoring intensity greater than 40, 50, and 60 dB were measured by Watch-PAT 100. Correlations of these parameters with apnea hypopnea index (AHI), respiratory disturbance index (RDI), and oxygen desaturation index were assessed.

Results

The mean age and body mass index were 46.5±14.8 years and 24.7±3.4 kg/m², respectively. Mean AHI and RDI were 16.5±15.3/hour and 20.8±14.3/hour, respectively. The mean snoring intensity in the no, mild, moderate, and severe OSA groups was 44.0±2.7, 45.4±6.0, 47.7±5.0, and 50.5±5.6 dB, respectively (P<0.001). There was a positive correlation between snoring intensity and pAHI or PAT RDI (pRDI) (r=0.391 and r=0.385, respectively, both P<0.001). There was also a positive correlation between percent sleep time with the snoring intensity greater than 50 dB and pAHI or pRDI (r=0.423 and r=0.411, respectively, both P<0.001).

Conclusion

This study revealed that the intensity of snoring increased with the severity of sleep apnea, which suggests that the loudness of snoring might be an indicator of the severity of OSA.

Hidden Markov modelling of intra-snore episode behavior of acoustic characteristics of obstructive sleep apnea patients

Obstructive sleep apnea (OSA) is a breathing disorder that can cause serious medical consequences. It is caused by full (apnea) or partial (hypopnea) obstructions of the upper airway during sleep. The gold standard for diagnosis of OSA is the polysomnography (PSG). The main measure for OSA diagnosis is the apnea-hypopnea index (AHI). However, the AHI is a time averaged summary measure of vast amounts of information gathered in an overnight PSG study. It cannot capture the dynamic characteristics associated with apnea/hypopnea events and their overnight distribution. The dynamic characteristics of apnea/hypopnea events are affected by the structural and functional characteristics of the upper airway. The upper airway characteristics also affect the upper airway collapsibility. These effects are manifested in snoring sounds generated from the vibrations of upper airway structures which are then modified by the upper airway geometric and physical characteristics. Hence, it is highly likely that the acoustical behavior of snoring is affected by the upper airway structural and functional characteristics. In the current work, we propose a novel method to model the intra-snore episode behavior of the acoustic characteristics of snoring sounds which can indirectly describe the instantaneous and temporal dynamics of the upper airway. We model the intra-snore episode acoustical behavior by using hidden Markov models (HMMs) with Mel frequency cepstral coefficients. Assuming significant differences in the anatomical and physiological upper airway configurations between low-AHI and high-AHI subjects, we defined different snorer groups with respect to AHI thresholds 15 and 30 and also developed HMM-based classifiers to classify snore episodes into those groups. We also define a measure called instantaneous apneaness score (IAS) in terms of the log-likelihoods produced by respective HMMs. IAS indicates the degree of class membership of each episode to one of the predefined groups as well as the instantaneous OSA severity. We then assigned each patient to an overall AHI band based on the majority vote of each episode of snoring. The proposed method has a diagnostic sensitivity and specificity between 87-91%.

OSA severity assessment based on sleep breathing analysis using ambient microphone

In this paper, an audio-based system for severity estimation of obstructive sleep apnea (OSA) is proposed. The system estimates the apnea-hypopnea index (AHI), which is the average number of apneic events per hour of sleep. This system is based on a Gaussian mixture regression algorithm that was trained and validated on full-night audio recordings. Feature selection process using a genetic algorithm was applied to select the best features extracted from time and spectra domains. A total of 155 subjects, referred to in-laboratory polysomnography (PSG) study, were recruited. Using the PSG's AHI score as a gold-standard, the performances of the proposed system were evaluated using a Pearson correlation, AHI error, and diagnostic agreement methods. Correlation of R=0.89, AHI error of 7.35 events/hr, and diagnostic agreement of 77.3% were achieved, showing encouraging performances and a reliable non-contact alternative method for OSA severity estimation.

Acoustic analysis of overnight consecutive snoring sounds by sound pressure levels

CONCLUSION

The sound pressure level (SPL) parameters, especially the A-weighted equivalent sound level (LAeq) and accumulative percentile sound level 10 (L10), were significantly different between simple snoring (SS) and obstructive sleep apnea-hypopnea syndrome (OSAHS). The apnea-hypopnea index (AHI) was the most significant factor to affect the SPLs of snoring sounds. LAeq and L10 were valuable acoustic characters of snoring which could reflect the severity of sleep disordered breathing in clinic.

OBJECTIVES

Due to the limitation of acoustic analysis of single snoring sound for snorers, this study analyzed characteristics of consecutive snoring sounds overnight by the SPLs in patients of SS and OSAHS.

METHOD

Ninety-four patients who underwent simultaneous SPL recording and polysomnography (PSG) were included in this study. Parameters of SPL such as LAeq, peak sound level (Lpeak), L10, L50, and L90 were analyzed. The correlation between these parameters and PSG results was also analyzed.

RESULTS

The LAeq and L10 in OSAHS patients were significantly different from patients with SS. The body mass index (BMI) was positively correlated to LAeq and L10. Among various factors of PSG data and demographic factors, the SPLs were mostly affected by the AHI and the lowest oxygen saturation (LSaO2).

Acoustic Analysis of Snoring in the Diagnosis of Obstructive Sleep Apnea Syndrome: A Call for More Rigorous Studies

BACKGROUND

Snoring is a common symptom of obstructive sleep apnea syndrome (OSA) and has recently been considered for diagnosis of OSA.

OBJECTIVES

The goal of the current study was to systematically determine the accuracy of acoustic analysis of snoring in the diagnosis of OSA using a meta-analysis.

METHODS

PubMed, Cochrane Library database, and EMBASE were searched up to July 15, 2014. A systematic review and meta-analysis of sensitivity, specificity, and other measures of accuracy of acoustic analysis of snoring in the diagnosis of OSA were conducted. The median of apneahypopnea index threshold was 10 events/h, range: 5-15 or 10-15 if aforementioned suggestion is adopted.

RESULTS

A total of seven studies with 273 patients were included in the meta-analysis. The pooled estimates were as follows: sensitivity, 88% (95% confidence interval [CI]: 82-93%); specificity, 81% (95% CI: 72-88%); positive likelihood ratio (PLR), 4.44 (95% CI: 2.39-8.27); negative likelihood ratio (NLR), 0.15 (95% CI: 0.10-0.24); and diagnostic odds ratio (DOR), 32.18 (95% CI: 13.96-74.81). χ(2) values of sensitivity, specificity, PLR, NLR, and DOR were 2.37, 10.39, 12.57, 3.79, and 6.91 respectively (All p > 0.05). The area under the summary receiver operating characteristic curve was 0.93. Sensitivity analysis demonstrated that the pooled estimates were stable and reliable. The results of publication bias were not significant (p = 0.30).

CONCLUSIONS

Acoustic analysis of snoring is a relatively accurate but not a strong method for diagnosing OSA. There is an urgent need for rigorous studies involving large samples and single snore event tests with an efficacy criterion that reflects the particular features of snoring acoustics for OSA diagnosis.

Is There a Relationship Between Snoring Sound Intensity and Frequency and OSAS Severity?

Objectives:

We investigated the relationship between snoring sounds and severity of obstructive sleep apnea syndrome (OSAS).

Methods:

A total number of 103 snoring patients (60 males and 43 females) were evaluated by means of polysomnographic findings and snoring sound recordings. Snoring sound intensity was assessed using fast Fourier transform (FFT) method by measuring maximal frequency (Fmax) and average snoring sound intensity level (SSIL).

Results:

Maximal frequency and SSIL are correlated with apnea-hypopnea index (AHI), REM AHI, and severity of the OSAS. So, as the severity of the OSAS increased, so did the Fmax and SSIL of the snoring recordings, meaning patients started snoring louder with more frequency. In older patients, in females, in severe OSAS group, and in patients with higher body mass index (BMI), AHI and AHI REM values and SSIL and Fmax values increased. As mean oxygen (O2) saturation and lowest O2 saturation decreased, SSIL and Fmax values increased.

Conclusion:

Maximal frequency and SSIL analysis of the snoring sound increased in severe OSAS patients. People should be aware of the importance of snoring sounds. In particular, patients with snoring sounds increasing in intensity and of higher frequency should discuss with their physicians the possibility of OSAS.

Synchronized imaging and acoustic analysis of the upper airway in patients with sleep-disordered breathing

Progressive narrowing of the upper airway increases airflow resistance and can produce snoring sounds and apnea/hypopnea events associated with sleep-disordered breathing due to airway collapse. Recent studies have shown that acoustic properties during snoring can be altered with anatomic changes at the site of obstruction. To evaluate the instantaneous association between acoustic features of snoring and the anatomic sites of obstruction, a novel method was developed and applied in nine patients to extract the snoring sounds during sleep while performing dynamic magnetic resonance imaging (MRI). The degree of airway narrowing during the snoring events was then quantified by the collapse index (ratio of airway diameter preceding and during the events) and correlated with the synchronized acoustic features. A total of 201 snoring events (102 pure retropalatal and 99 combined retropalatal and retroglossal events) were recorded, and the collapse index as well as the soft tissue vibration time were significantly different between pure retropalatal (collapse index, 2 ± 11%; vibration time, 0.2 ± 0.3 s) and combined (retropalatal and retroglossal) snores (collapse index, 13 ± 7% [P ≤ 0.0001]; vibration time, 1.2 ± 0.7 s [P ≤ 0.0001]). The synchronized dynamic MRI and acoustic recordings successfully characterized the sites of obstruction and established the dynamic relationship between the anatomic site of obstruction and snoring acoustics.

Nocturnal snoring sound analysis in the diagnosis of obstructive sleep apnea in the Chinese Han population

PURPOSE

Loud snoring is one of the principle symptoms of obstructive sleep apnea (OSA). Snoring sound analysis is a potentially cost-effective, reliable alternative for the diagnosis of OSA. However, no investigation has determined the accuracy of snoring signal analysis for the diagnosis of OSA in the Chinese Han population. Therefore, we investigated whether whole-night snoring detection and analysis aids the diagnosis of OSA using a new snore analysis technique.

METHODS

Snoring sounds were recorded using a non-contact microphone and polysomnography (PSG) was performed simultaneously throughout the night. We randomly selected 30 subjects each from four groups based on the severity of OSA. The rhythm and frequency domain of the snoring signal were analyzed based on frequency energy endpoint detection (FEP) and the Earth mover's distance (EMD), for each subject to harvest the EMD-calculated Apnea-Hypopnea Index (AHIEMD). Finally, we compared the AHIEMD with the PSG-monitored AHI (AHIPSG).

RESULTS

The accuracy of the AHIEMD compared with the AHIPSG was 96.7, 86.7, 86.7, and 96.7% in non-, mild, moderate, and severe OSA patients, respectively. AHIEMD was correlated with AHIPSG (r(2) = 0.950, p < 0.001). The area under the receiver operating characteristic curve values for OSA detection was 0.974, 0.957, and 0.997 for AHIEMD thresholds of 5, 15, and 30 events/h, respectively. Bland-Altman analysis revealed 91.7% agreement of AHIEMD with AHIPSG.

CONCLUSIONS

This new method for identifying OSA by analyzing snoring is feasible and reliable in the Han population. The snoring sound-based technique appears to be a promising tool for OSA screening and diagnosis.

Telephonic Analysis of the Snoring Sound Spectrum

Objective:

Snoring is a sound caused by vibration of collapsed and/or unsteady airway walls of the pharynx and soft palate. We compared stored spectra of snoring sounds recorded via cell phone (CP) and a microphone placed over the head (head phone [HP]).

Methods:

Thirty-four snoring patients were included in this prospective study. Groups were identified by reference to body mass index (BMI) values: group 1, BMI < 25 kg/m2 (n = 8); group 2, BMI 25 to 29 kg/m2 (n = 10); and group 3, BMI ≥ 30 kg/m2 (n = 16). Snoring sounds were recorded using CPs and HPs and digitally analyzed. We identified the frequencies with the highest snoring powers (Fmax values) and snoring sound intensity levels (SSILs).

Results:

Fmax ranged from 520 to 985 Hz in HP recordings and from 845 to 1645 Hz in CP recordings. Snoring sound intensity level values increased in proportion to BMI and were 6 to 24 dB in HP recordings and 19 to 52 dB in CP recordings. Thus, the CP values of Fmax and SSIL were higher than the HP values. In obese patients of group 3, almost all Fmax and SSIL values were higher than those of groups 1 and 2. In particular, the CP Fmax values were elevated in such patients. The advanced technologies used in modern CPs may allow some snoring sounds in susceptible individuals to be defined as oronasal.

Conclusion:

Cell phone technology allows snoring to be evaluated in patients located in areas remote from a hospital. To explore the intensity of snoring and to postoperatively monitor the efficacy of surgery used to treat snoring, telephonic sound analysis is both new and effective and reduces the need for patient attendance at a hospital. Those experiencing severe snoring and/or who are obese should be told of what can be done to solve such problems.

Simple snoring: not quite so simple after all?

Simple snoring (SS), in the absence of obstructive sleep apnoea (OSA), is a common problem, yet our understanding of its causes and consequences is incomplete. Our understanding is blurred by the lack of consistency in the definition of snoring, methods of assessment, and degree of concomitant complaints. Further, it remains contentious whether SS is independently associated with daytime sleepiness, or adverse health outcomes including cardiovascular disease and metabolic syndrome. Regardless of this lack of clarity, it is likely that SS exists on one end of a continuum, with OSA at its polar end. This possibility highlights the necessity of considering an otherwise 'annoying' complaint, as a serious risk factor for the development and progression of sleep apnoea, and consequent poor health outcomes. In this review, we: 1) highlight variation in prevalence estimates of snoring; 2) review the literature surrounding the distinctions between SS, upper airway resistance syndrome (UARS) and OSA; 3) present the risk factors for SS, in as far as it is distinguishable from UARS and OSA; and 4) describe common correlates of snoring, including cardiovascular disease, metabolic syndrome, and daytime sleepiness.

Capability of a neck worn device to measure sleep/wake, airway position, and differentiate benign snoring from obstructive sleep apnea

To evaluate the accuracy of a neck-worn device in measuring sleep/wake, detecting supine airway position, and using loud snoring to screen for obstructive sleep apnea. Study A included 20 subjects who wore the neck-device during polysomnography (PSG), with 31 records obtained from diagnostic and split-night studies. Study B included 24 community-based snorers studied in-home for up to three-nights with obstructive sleep apnea (OSA) severity measured with a validated Level III recorder. The accuracy of neck actigraphy-based sleep/wake was measured by assessing sleep efficiency (SE). Differences in sleep position measured at the chest and neck during PSG were compared to video-editing. Loud snoring acquired with an acoustic microphone was compared to the apnea-hypopnea index (AHI) by- and acrosspositions. Over-reported SE by neck actigraphy was inversely related to OSA severity. Measurement of neck and chest supine position were highly correlated with video-edits (r = 0.93, 0.78). Chest was bias toward over-estimating supine time while the majority of neck-device supine position errors occurred during CPAP titrations. Snoring was highly correlated with the overall, supine, and non-supine PSG-AHI (r = 0.79, 0.74, 0.83) and was both sensitive and specific in detecting overall, supine, and non-supine PSGAHI >10 (sensitivity = 81, 88, 82 %; specificity = 87, 79, 100 %). At home sleep testing-AHI > 10, the sensitivity and specificity of loud snoring was superior when users were predominantly non-supine as compared to baseline (sensitivity = 100, 92 %; specificity = 88, 77 %). Neck actigraphy appears capable of estimating sleep/wake. The accuracy of supine airway detection with the neck-device warrants further investigation. Measurement of loud snoring appears to provide a screening tool for differentiating positional apneic and benign snorers.

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

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.

A review of signals used in sleep analysis

This article presents a review of signals used for measuring physiology and activity during sleep and techniques for extracting information from these signals. We examine both clinical needs and biomedical signal processing approaches across a range of sensor types. Issues with recording and analysing the signals are discussed, together with their applicability to various clinical disorders. Both univariate and data fusion (exploiting the diverse characteristics of the primary recorded signals) approaches are discussed, together with a comparison of automated methods for analysing sleep.

Automatic detection of whole night snoring events using non-contact microphone

Objective

Although awareness of sleep disorders is increasing, limited information is available on whole night detection of snoring. Our study aimed to develop and validate a robust, high performance, and sensitive whole-night snore detector based on non-contact technology.

Design

Sounds during polysomnography (PSG) were recorded using a directional condenser microphone placed 1 m above the bed. An AdaBoost classifier was trained and validated on manually labeled snoring and non-snoring acoustic events.

Patients

Sixty-seven subjects (age 52.5±13.5 years, BMI 30.8±4.7 kg/m², m/f 40/27) referred for PSG for obstructive sleep apnea diagnoses were prospectively and consecutively recruited. Twenty-five subjects were used for the design study; the validation study was blindly performed on the remaining forty-two subjects.

Measurements and Results

To train the proposed sound detector, >76,600 acoustic episodes collected in the design study were manually classified by three scorers into snore and non-snore episodes (e.g., bedding noise, coughing, environmental). A feature selection process was applied to select the most discriminative features extracted from time and spectral domains. The average snore/non-snore detection rate (accuracy) for the design group was 98.4% based on a ten-fold cross-validation technique. When tested on the validation group, the average detection rate was 98.2% with sensitivity of 98.0% (snore as a snore) and specificity of 98.3% (noise as noise).

Conclusions

Audio-based features extracted from time and spectral domains can accurately discriminate between snore and non-snore acoustic events. This audio analysis approach enables detection and analysis of snoring sounds from a full night in order to produce quantified measures for objective follow-up of patients.

Evaluation of position dependency in non-apneic snorers

The aims of this study are to determine the prevalence of position dependency in non-apneic snorers, as defined by the American Academy of Sleep Medicine (AASM) guidelines, and to investigate the influence of various factors such as BMI, neck circumference, age, gender, and sleep efficiency on sleeping position. A cohort of consecutive patients was screened for complaints of excessive snoring or symptoms suspicious for sleep disordered breathing. Overnight polysomnographic data were collected and non-apneic snorers who met all the inclusion criteria were selected for statistical analysis. To assess position-dependent snoring, the snore index (total snores/h) was used. Supine-dependent patients were defined as having a supine snore index higher than their total non-supine snore index. 76 patients were eligible for statistical analysis. Prevalence of position dependency in non-apneic snorers was 65.8% (p < 0.008). A stepwise regression showed that only BMI had a significant effect (p < 0.003) on the supine snore index. This is the first study that uses the AASM guidelines to accurately define non-apneic snorers (AHI < 5) and provides scientific evidence that the majority of non-apneic snorers are supine dependent. Furthermore, these results show that non-apneic snorers with a higher BMI snore more frequently in supine position. The use of sleep position therapy therefore, has the potential to play a significant role in improving snoring and its associated physical and psychosocial health outcomes in this population.

Energy types of snoring sounds in patients with obstructive sleep apnea syndrome: a preliminary observation

Background

Annoying snore is the principle symptom and problem in obstructive sleep apnea syndrome (OSAS). However, investigation has been hampered by the complex snoring sound analyses.

Objective

This study was aimed to investigate the energy types of the full-night snoring sounds in patients with OSAS.

Patients and Method

Twenty male OSAS patients underwent snoring sound recording throughout 6 hours of in-lab overnight polysomnogragphy. Snoring sounds were processed and analyzed by a new sound analytic program, named as Snore Map®. We transformed the 6-hour snoring sound power spectra into the energy spectrum and classified it as snore map type 1 (monosyllabic low-frequency snore), type 2 (duplex low-&mid-frequency snore), type 3 (duplex low- & high-frequency snore), and type 4 (triplex low-, mid-, & high-frequency snore). The interrator and test-retest reliabilities of snore map typing were assessed. The snore map types and their associations among demographic data, subjective snoring questionnaires, and polysomnographic parameters were explored.

Results

The interrator reliability of snore map typing were almost perfect (κ = 0.87) and the test-retest reliability was high (r = 0.71). The snore map type was proportional to the body mass index (r = 0.63, P = 0.003) and neck circumference (r = 0.52, P = 0.018). Snore map types were unrelated to subjective snoring questionnaire scores (All P>0.05). After adjustment for body mass index and neck circumference, snore map type 3–4 was significantly associated with severity of OSAS (r = 0.52, P = 0.026).

Conclusions

Snore map typing of a full-night energy spectrum is feasible and reliable. The presence of a higher snore map type is a warning sign of severe OSAS and indicated priority OSAS management. Future studies are warranted to evaluate whether snore map type can be used to discriminate OSAS from primary snoring and whether it is affected by OSAS management.

Obstructive apnea hypopnea index estimation by analysis of nocturnal snoring signals in adults

STUDY OBJECTIVE

To develop a whole-night snore sounds analysis algorithm enabling estimation of obstructive apnea hypopnea index (AHI(EST)) among adult subjects.

DESIGN

Snore sounds were recorded using a directional condenser microphone placed 1 m above the bed. Acoustic features exploring intra-(mel- cepstability, pitch density) and inter-(running variance, apnea phase ratio, inter-event silence) snore properties were extracted and integrated to assess AHI(EST).

SETTING

University-affiliated sleep-wake disorder center and biomedical signal processing laboratory.

PATIENTS

Ninety subjects (age 53 ± 13 years, BMI 31 ± 5 kg/m(2)) referred for polysomnography (PSG) diagnosis of OSA were prospectively and consecutively recruited. The system was trained and tested on 60 subjects. Validation was blindly performed on the additional 30 consecutive subjects.

MEASUREMENTS AND RESULTS

AHI(EST) correlated with AHI (AHI(PSG); r(2) = 0.81, P < 0.001). Area under the receiver operating characteristic curve of 85% and 92% for thresholds of 10 and 20 events/h, respectively, were obtained for OSA detection. Both Altman-Bland analysis and diagnostic agreement criteria revealed 80% and 83% agreements of AHI(EST) with AHI(PSG), respectively.

CONCLUSIONS

Acoustic analysis based on intra- and inter-snore properties can differentiate subjects according to AHI. An acoustic-based screening system may address the growing needs for reliable OSA screening tool. Further studies are needed to support these findings.

Normal non-regular snores as a tool for screening SAHS severity

Snoring is one of the earliest and most consistent sign of upper airway obstruction leading to Sleep Apnea-Hypopnea Syndrome (SAHS). Several studies on post-apneic snores, snores that are emitted immediately after an apnea, have already proven that this type of snoring is most distinct from that of normal snoring. However, post-apneic snores are more unlikely and sometimes even inexistent in simple snorers and mild SAHS subjects. In this work we address that issue by proposing the study of normal non-regular snores. They correspond to successive snores that are separated by normal breathing cycles. The results obtained establish the feasibility of acoustic parameters of normal non-regular snores as a promising tool for a prompt screening of SAHS severity.

Bayes classification of snoring subjects with and without Sleep Apnea Hypopnea Syndrome, using a Kernel method

The gold standard for diagnosing Sleep Apnea Hypopnea Syndrome (SAHS) is the Polysomnography (PSG), an expensive, labor-intensive and time-consuming procedure. It would be helpful to have a simple screening method that allowed to early determining the severity of a subject prior to his/her enrolment for a PSG. Several differences have been reported in the acoustic snoring characteristics between simple snorers and SAHS patients. Previous studies usually classify snoring subjects into two groups given a threshold of Apnea-Hypoapnea Index (AHI). Recently, Bayes multi-group classification with Gaussian Probability Density Function (PDF) has been proposed, using snore features in combination with apnea-related information. In this work we show that the Bayes classifier with Kernel PDF estimation outperforms the Gaussian approach and allows the classification of SAHS subjects according to their severity, using only the information obtained from snores. This could be the base of a single channel, snore-based, screening procedure for SAHS.

Snoring analysis for the screening of Sleep Apnea Hypopnea Syndrome with a single-channel device developed using polysomnographic and snoring databases

Several studies have shown differences in acoustic snoring characteristics between patients with Sleep Apnea-Hypopnea Syndrome (SAHS) and simple snorers. Usually a few manually isolated snores are analyzed, with an emphasis on postapneic snores in SAHS patients. Automatic analysis of snores can provide objective information over a longer period of sleep. Although some snore detection methods have recently been proposed, they have not yet been applied to full-night analysis devices for screening purposes. We used a new automatic snoring detection and analysis system to monitor snoring during full-night studies to assess whether the acoustic characteristics of snores differ in relation to the Apnea-Hypopnea Index (AHI) and to classify snoring subjects according to their AHI. A complete procedure for device development was designed, using databases with polysomnography (PSG) and snoring signals. This included annotation of many types of episodes by an expert physician: snores, inspiration and exhalation breath sounds, speech and noise artifacts, The AHI of each subject was estimated with classical PSG analysis, as a gold standard. The system was able to correctly classify 77% of subjects in 4 severity levels, based on snoring analysis and sound-based apnea detection. The sensitivity and specificity of the system, to identify healthy subjects from pathologic patients (mild to severe SAHS), were 83% and 100%, respectively. Besides, the Apnea Index (AI) obtained with the system correlated with the obtained by PSG or Respiratory Polygraphy (RP) (r=0.87, p<0.05).

All night analysis of time interval between snores in subjects with sleep apnea hypopnea syndrome

Sleep apnea–hypopnea syndrome (SAHS) is a serious sleep disorder, and snoring is one of its earliest and most consistent symptoms. We propose a new methodology for identifying two distinct types of snores: the so-called non-regular and regular snores. Respiratory sound signals from 34 subjects with different ranges of Apnea-Hypopnea Index (AHI = 3.7–109.9 h⁻¹) were acquired. A total number of 74,439 snores were examined. The time interval between regular snores in short segments of the all night recordings was analyzed. Severe SAHS subjects show a shorter time interval between regular snores (p = 0.0036, AHI cp: 30 h⁻¹) and less dispersion on the time interval features during all sleep. Conversely, lower intra-segment variability (p = 0.006, AHI cp: 30 h⁻¹) is seen for less severe SAHS subjects. Features derived from the analysis of time interval between regular snores achieved classification accuracies of 88.2 % (with 90 % sensitivity, 75 % specificity) and 94.1 % (with 94.4 % sensitivity, 93.8 % specificity) for AHI cut-points of severity of 5 and 30 h⁻¹, respectively. The features proved to be reliable predictors of the subjects’ SAHS severity. Our proposed method, the analysis of time interval between snores, provides promising results and puts forward a valuable aid for the early screening of subjects suspected of having SAHS.

Multiclass classification of subjects with sleep apnoea-hypopnoea syndrome through snoring analysis

The gold standard for diagnosing sleep apnoea-hypopnoea syndrome (SAHS) is polysomnography (PSG), an expensive, labour-intensive and time-consuming procedure. Accordingly, it would be very useful to have a screening method to allow early assessment of the severity of a subject, prior to his/her referral for PSG. Several differences have been reported between simple snorers and SAHS patients in the acoustic characteristics of snoring and its variability. In this paper, snores are fully characterised in the time domain, by their sound intensity and pitch, and in the frequency domain, by their formant frequencies and several shape and energy ratio measurements. We show that accurate multiclass classification of snoring subjects, with three levels of SAHS, can be achieved on the basis of acoustic analysis of snoring alone, without any requiring information on the duration or the number of apnoeas. Several classification methods are examined. The best of the approaches assessed is a Bayes model using a kernel density estimation method, although good results can also be obtained by a suitable combination of two binary logistic regression models. Multiclass snore-based classification allows early stratification of subjects according to their severity. This could be the basis of a single channel, snore-based screening procedure for SAHS.

Regular and non regular snore features as markers of SAHS

Sleep Apnea-Hypopnea Syndrome (SAHS) diagnosis is still done with an overnight multi-channel polysomnography. Several efforts are being made to study profoundly the snore mechanism and discover how it can provide an opportunity to diagnose the disease. This work introduces the concept of regular snores, defined as the ones produced in consecutive respiratory cycles, since they are produced in a regular way, without interruptions. We applied 2 thresholds (TH(adaptive) and TH(median)) to the time interval between successive snores of 34 subjects in order to select regular snores from the whole all-night snore sequence. Afterwards, we studied the effectiveness that parameters, such as time interval between successive snores and the mean intensity of snores, have on distinguishing between different levels of SAHS severity (AHI (Apnea-Hypopnea Index) < 5h(-1), AHI <10 h(-1), AHI < 15 h(-1), AHI < 30 h(-1)). Results showed that TH(adaptive) outperformed TH(median) on selecting regular snores. Moreover, the outcome achieved with non-regular snores intensity features suggests that these carry key information on SAHS severity.

Nocturnal sound analysis for the diagnosis of obstructive sleep apnea

A novel method for screening obstructive sleep apnea syndrome (OSAs) based on nocturnal acoustic signal is proposed. Full-night audio signals from sixty subjects were segmented into snore, noise and silence events using semi-automatic algorithm based on Gaussian mixture models which achieves more than 90% (92%) sensitivity (specificity) and produces an average of 2,000 snores per subject. A classification into 3 groups is proposed for the diagnosis: comparison group - non-OSA subjects (apnea hypopnea index, AHI < 10), mild to moderate OSA (10 < AHI < 30) and severe OSA (AHI>30). A Bayes classifier was implemented, fed with five acoustic features, all correlated with the severity of the syndrome: (1) Inter Event Silence, which quantifies segments suspicious as apnea; (2) Mel Cepstability, measures the entire night stability of the spectrum, expressed using mel-frequency cepstrum; (3) Energy Running Variance, a criterion for the variation of the nocturnal acoustic pattern; (4) Apneic Phase Ratio, exploiting the finding that snores around apnea events expressing larger acoustic variation; and (5) Pitch Density. Correct classification of 92% for resubstitution method and 80% for 5-fold cross validation method was achieved. Moreover, in a case of two groups with a threshold of AHI=10, a sensitivity (specificity) of 96.5% (90.6%) and 87.5% (82.1%) for resubstitution and cross-validation respectively were obtained.