Comparison of SVM and ANFIS for snore related sounds classification by using the largest Lyapunov exponent and entropy

Using entropy of snoring, respiratory effort and electrocardiography signals during sleep for OSA detection and severity classification

STUDY OBJECTIVES

Obstructive sleep apnea (OSA) is a very prevalent disease and its diagnosis is based on polysomnography (PSG). We investigated whether snoring-sound-, very low frequency electrocardiogram (ECG-VLF)- and thoraco-abdominal effort- PSG signal entropy values could be used as surrogate markers for detection of OSA and OSA severity classification.

METHODS

The raw data of the snoring-, ECG- and abdominal and thoracic excursion signal recordings of two consecutive full-night PSGs of 86 consecutive patients (22 female, 53.74 ± 12.4 years) were analyzed retrospectively. Four epochs (30 s each, manually scored according to the American Academy of Sleep Medicine standard) of each sleep stage (N1, N2, N3, REM, awake) were used as the ground truth. Sampling entropy (SampEn) of all the above signals was calculated and group comparisons between the OSA severity groups were performed. In total, (86x4x5 = )1720 epochs/group/night were included in the training set as an input for a support vector machine (SVM) algorithm to classify the OSA severity classes. Analyses were performed for first- and second-night PSG recordings separately.

RESULTS

Twenty-seven patients had mild (RDI = ≥ 5/h but <15/h), 21 patients moderate (RDI ≥15/h but <30/h) and 23 patients severe OSA (RDI ≥30/h). Fifteen patients had an RDI <5/h and were therefore considered non-OSA. Using SE on the above three PSG signal data and using a SVM pipeline, it was possible to distinguish between the four OSA severity classes. The best metric was snoring signal-SE. The area-under-the-curve (AUC) calculations showed reproducible significant results for both nights of PSG. The second night data were even more significant, with non-OSA (R) vs. light OSA (L) 0.61, R vs. moderate (M) 0.68, R vs. heavy OSA (H) 0.84, L vs. M 0.63, M vs. H 0.65 and L vs. H 0.82. The results were not confounded by age or gender.

CONCLUSIONS

SampEn of either snoring-, very low ECG-frequencies- or thoraco-abdominal effort signals alone may be used as a surrogate marker to diagnose OSA and even predict OSA severity. More specifically, in this exploratory study snoring signal SampEn showed the greatest predictive accuracy for OSA among the three signals. Second night data showed even more accurate results for all three parameters than first-night recordings. Therefore, technologies using only parts of the PSG signal, e.g. sound-recording devices, may be used for OSA screening and OSA severity group classification.

Estimating a Sleep Apnea Hypopnea Index Based on the ERB Correlation Dimension of Snore Sounds

This paper proposes a new perspective of analyzing non-linear acoustic characteristics of the snore sounds. According to the ERB (Equivalent Rectangular Bandwidth) scale used in psychoacoustics, the ERB correlation dimension (ECD) of the snore sound was computed to feature different severity levels of sleep apnea hypopnea syndrome (SAHS). For the training group of 93 subjects, snore episodes were manually segmented and the ECD parameters of the snores were extracted, which established the gaussian mixture models (GMM). The nocturnal snore sound of the testing group of another 120 subjects was tested to detect SAHS snores, thus estimating the apnea hypopnea index (AHI), which is called AHIECD. Compared to the AHIPSG value of the gold standard polysomnography (PSG) diagnosis, the estimated AHIECD achieved an accuracy of 87.5% in diagnosis the SAHS severity levels. The results suggest that the ECD vectors can be effective parameters for screening SAHS.

Overnight airway obstruction severity prediction centered on acoustic properties of smart phone: validation with esophageal pressure

OBJECTIVE

Obstructive sleep apnea is characterized by a number of airway obstructions. Esophageal pressure manometry (EPM) based estimation of consecutive peak to trough differences (ΔPes) is the gold standard method to quantify the severity of airway obstructions. However, the procedure is rarely available in sleep laboratories due to invasive nature. There is a clinical need for a simplified, scalable technology that can quantify the severity of airway obstructions. In this paper, we address this and propose a pioneering technology, centered on sleep related respiratory sound (SRS) to predict overnight ΔPes signal.

APPROACH

We recorded streams of SRS using a bedside iPhone 7 smartphone from subjects undergoing diagnostic polysomnography (PSG) studies and EPM was performed concurrently. Overnight data was divided into epochs of 10 s duration with 50% overlap. Altogether, we extracted 42 181 such epochs from 13 subjects. Acoustic features and features from the two PSG signals serve as an input to train a machine learning algorithm to achieve mapping between non-invasive features and ΔPes values. A testing dataset of 14 171 epochs from four new subjects was used for validation.

MAIN RESULTS

The SRS based model predicted the ΔPes with a median of absolute error of 6.75 cmH2O (±0.59, r = 0.83(±0.03)). When information from the PSG were combined with the SRS, the model performance became: 6.37cmH2O (±1.02, r = 0.85(±0.04)).

SIGNIFICANCE

The smart phone based SRS alone, or in combination with routinely collected PSG signals can provide a non-invasive method to predict overnight ΔPes. The method has the potential to be automated and scaled to provide a low-cost alternative to EPM.

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.

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.

Synchronization of coupled different chaotic FitzHugh-Nagumo neurons with unknown parameters under communication-direction-dependent coupling

This paper investigates the chaotic behavior and synchronization of two different coupled chaotic FitzHugh-Nagumo (FHN) neurons with unknown parameters under external electrical stimulation (EES). The coupled FHN neurons of different parameters admit unidirectional and bidirectional gap junctions in the medium between them. Dynamical properties, such as the increase in synchronization error as a consequence of the deviation of neuronal parameters for unlike neurons, the effect of difference in coupling strengths caused by the unidirectional gap junctions, and the impact of large time-delay due to separation of neurons, are studied in exploring the behavior of the coupled system. A novel integral-based nonlinear adaptive control scheme, to cope with the infeasibility of the recovery variable, for synchronization of two coupled delayed chaotic FHN neurons of different and unknown parameters under uncertain EES is derived. Further, to guarantee robust synchronization of different neurons against disturbances, the proposed control methodology is modified to achieve the uniformly ultimately bounded synchronization. The parametric estimation errors can be reduced by selecting suitable control parameters. The effectiveness of the proposed control scheme is illustrated via numerical simulations.

The generalization error bound for the multiclass analytical center classifier

This paper presents the multiclass classifier based on analytical center of feasible space (MACM). This multiclass classifier is formulated as quadratic constrained linear optimization and does not need repeatedly constructing classifiers to separate a single class from all the others. Its generalization error upper bound is proved theoretically. The experiments on benchmark datasets validate the generalization performance of MACM.