Utility of approximate entropy from overnight pulse oximetry data in the diagnosis of the obstructive sleep apnea syndrome

Approximate entropy (ApEn) is a family of statistics introduced as a quantification of regularity in time series without any a priori knowledge about the system generating them. The aim of this preliminary study was to assess whether a time series analysis of arterial oxygen saturation (SaO2) signals from overnight pulse oximetry by means of ApEn could yield essential information on the diagnosis of obstructive sleep apnea (OSA) syndrome. We analyzed SaO2 signals from 187 subjects: 111 with a positive diagnosis of OSA and 76 with a negative diagnosis of OSA. We divided our data in a training set (44 patients with OSA Positive and 30 patients with OSA Negative) and a test set (67 patients with OSA Positive and 46 patients with OSA Negative). The training set was used for algorithm development and optimum threshold selection. Results showed that recurrence of apnea events in patients with OSA determined a significant increase in ApEn values. This method was assessed prospectively using the test dataset, where we obtained 82.09% sensitivity and 86.96% specificity. We conclude that ApEn analysis of SaO2 from pulse oximetric recording could be useful in the study of OSA.     

Automatic detection of obstructive sleep apnea using speech signals

Obstructive sleep apnea (OSA) is a common disorder associated with anatomical abnormalities of the upper airways that affects 5% of the population. Acoustic parameters may be influenced by the vocal tract structure and soft tissue properties. We hypothesize that speech signal properties of OSA patients will be different than those of control subjects not having OSA. Using speech signal processing techniques, we explored acoustic speech features of 93 subjects who were recorded using a text-dependent speech protocol and a digital audio recorder immediately prior to polysomnography study. Following analysis of the study, subjects were divided into OSA (n=67) and non-OSA (n=26) groups. A Gaussian mixture model-based system was developed to model and classify between the groups; discriminative features such as vocal tract length and linear prediction coefficients were selected using feature selection technique. Specificity and sensitivity of 83% and 79% were achieved for the male OSA and 86% and 84% for the female OSA patients, respectively. We conclude that acoustic features from speech signals during wakefulness can detect OSA patients with good specificity and sensitivity. Such a system can be used as a basis for future development of a tool for OSA screening.     

Spectral and nonlinear analyses of MEG background activity in patients with Alzheimer's disease.

The aim of the present study is to analyze the magnetoencephalogram (MEG) background activity from patients with Alzheimer's disease (AD) and elderly control subjects. MEG recordings from 20 AD patients and 21 controls were analyzed by means of two spectral [median frequency (MF) and spectral entropy (SpecEn)] and two nonlinear parameters [approximate entropy (ApEn) and Lempel-Ziv complexity (LZC)]. In the AD diagnosis, the highest accuracy of 75.6% (80% sensitivity, 71.4% specificity) was obtained with the MF according to a linear discriminant analysis (LDA) with a leave-one-out cross-validation procedure. Moreover, we wanted to assess whether these spectral and nonlinear analyses could provide complementary information to improve the AD diagnosis. After a forward stepwise LDA with a leave-one-out cross-validation procedure, one spectral (MF) and one nonlinear parameter (ApEn) were automatically selected. In this model, an accuracy of 80.5% (80.0% sensitivity, 81.0% specificity) was achieved. We conclude that spectral and nonlinear analyses from spontaneous MEG activity could be complementary methods to help in AD detection.     

Spectral and Nonlinear Analyses of MEG Background Activity in Patients With Alzheimer's Disease

The aim of the present study is to analyze the magnetoencephalogram (MEG) background activity from patients with Alzheimer's disease (AD) and elderly control subjects. MEG recordings from 20 AD patients and 21 controls were analyzed by means of two spectral [median frequency (MF) and spectral entropy (SpecEn)] and two nonlinear parameters [approximate entropy (ApEn) and Lempel-Ziv complexity (LZC)]. In the AD diagnosis, the highest accuracy of 75.6% (80% sensitivity, 71.4% specificity) was obtained with the MF according to a linear discriminant analysis (LDA) with a leave-one-out cross-validation procedure. Moreover, we wanted to assess whether these spectral and nonlinear analyses could provide complementary information to improve the AD diagnosis. After a forward stepwise LDA with a leave-one-out cross-validation procedure, one spectral (MF) and one nonlinear parameter (ApEn) were automatically selected. In this model, an accuracy of 80.5% (80.0% sensitivity, 81.0% specificity) was achieved. We conclude that spectral and nonlinear analyses from spontaneous MEG activity could be complementary methods to help in AD detection.     

Comparison of detrended fluctuation analysis and spectral analysis for heart rate variability in sleep and sleep apnea

Sleep has been regarded as a testing situation for the autonomic nervous system, because its activity is modulated by sleep stages. Sleep-related breathing disorders also influence the autonomic nervous system and can cause heart rate changes known as cyclical variation. We investigated the effect of sleep stages and sleep apnea on autonomic activity by analyzing heart rate variability (HRV). Since spectral analysis is suited for the identification of cyclical variations and detrended fluctuation analysis can analyze the scaling behavior and detect long-range correlations, we compared the results of both complementary techniques in 14 healthy subjects, 33 patients with moderate, and 31 patients with severe sleep apnea. The spectral parameters VLF, LF, HF, and LF/HF confirmed increasing parasympathetic activity from wakefulness and REM over light sleep to deep sleep, which is reduced in patients with sleep apnea. Discriminance analysis was used on a person and sleep stage basis to determine the best method for the separation of sleep stages and sleep apnea severity. Using spectral parameters 69.7% of the apnea severity assignments and 54.6% of the sleep stage assignments were correct, while using scaling analysis these numbers increased to 74.4% and 85.0%, respectively. We conclude that changes in HRV are better quantified by scaling analysis than by spectral analysis.     

Automatic control of airway pressure for treatment of obstructivesleep apnea

Obstructive sleep apnea (OSA) occurs when airflow ceases because of pharyngeal wall collapse in sleep. Repeated apneic events results in the development of a pathological condition called OSA syndrome. The authors describe the methodology and design of a prosthetic device, named automatic positive airway pressure (APAP), for treatment of this syndrome. HPAP applies a stream of air via a nasal mask at an initial pressure selected by the patient. By sensing specific pressure characteristics of air flow immediately preceding pharyngeal wall collapse, the APAP device automatically raises the applied pressure to maintain a patent upper airway and thus prevent apnea. Conversely, when such conditions are absent, pressure is lowered step wise until a preselected minimum pressure is reached. Performance evaluation of the APAP system in five OSA patients and five normal (asymptomatic for sleep apnea) subjects revealed that it effectively treated OSA syndrome. It lowered the apnea-hypopnea index without disturbing sleep and resulted in a lower mean airway pressure compared to the traditional continuous positive airway pressure (CPAP) therapy. The results also show that the pressure needed to prevent OSA varied significantly throughout the night. For OSA syndrome patients, this pressure ranged from 3 to 18 cm H ₂O. The mean airway pressure for these patients had a sample average of 6.80 cm H₂O and a standard deviation of 3.17 cm H ₂O. In normal subjects, the device did not raise pressure except in response to pharyngeal wall vibration events     

Midsagittal Jaw Movements as a Sleep/Wake Marker

The seriousness of the obstructive sleep apnea/hypopnea syndrome is measured by the apnea-hypopnea index (AHI), the number of sleep apneas and hypopneas over the total sleep time (TST). Cardiorespiratory signals are used to detect respiratory events while the TST is usually assessed by the analysis of electroencephalogram traces in polysomnography (PSG) or wrist actigraphy trace in portable monitoring. This paper presents a sleep/wake automatic detector that relies on a wavelet-based complexity measure of the midsagittal jaw movement signal and multilayer perceptrons. In all, 63 recordings were used to train and test the method, while 38 recordings constituted an independent evaluation set for which the sensitivity, the specificity, and the global agreement of sleep recognition, respectively, reached 85.1%, 76.4%, and 82.9%, compared with the PSG data. The AHI computed automatically and only from the jaw movement analysis was significantly improved (p < 0.0001 ) when considering this sleep/wake detector. Moreover, a sensitivity of 88.6% and a specificity of 83.6% were found for the diagnosis of the sleep apnea syndrome according to a threshold of 15. Thus, the jaw movement signal is reasonably accurate in separating sleep from wake, and, in addition to its ability to score respiratory events, is a valuable signal for portable monitoring.     

Automated recognition of obstructive sleep apnea syndrome using support vector machine classifier

Obstructive sleep apnea (OSA) is a common sleep disorder that causes pauses of breathing due to repetitive obstruction of the upper airways of the respiratory system. The effect of this phenomenon can be observed in other physiological signals like the heart rate variability, oxygen saturation, and the respiratory effort signals. In this study, features from these signals were extracted from 50 control and 50 OSA patients from the Sleep Heart Health Study database and implemented for minute and subject classifications. A support vector machine (SVM) classifier was used with linear and second-order polynomial kernels. For the minute classification, the respiratory features had the highest sensitivity while the oxygen saturation gave the highest specificity. The polynomial kernel always had better performance and the highest accuracy of 82.4% (Sen: 69.9%, Spec: 91.4%) was achieved using the combined-feature classifier. For subject classification, the polynomial kernel had a clear improvement in the oxygen saturation accuracy as the highest accuracy of 95% was achieved by both the oxygen saturation (Sen: 100%, Spec: 90.2%) and the combined-feature (Sen: 91.8%, Spec: 98.0%). Further analysis of the SVM with other kernel types might be useful for optimizing the classifier with the appropriate features for an OSA automated detection algorithm.     

Support Vector Machines for Automated Recognition of Obstructive Sleep Apnea Syndrome From ECG Recordings

Obstructive sleep apnea syndrome (OSAS) is associated with cardiovascular morbidity as well as excessive daytime sleepiness and poor quality of life. In this study, we apply a machine learning technique [support vector machines (SVMs)] for automated recognition of OSAS types from their nocturnal ECG recordings. A total of 125 sets of nocturnal ECG recordings acquired from normal subjects (OSAS- ) and subjects with OSAS (OSAS+), each of approximately 8 h in duration, were analyzed. Features extracted from successive wavelet coefficient levels after wavelet decomposition of signals due to heart rate variability (HRV) from RR intervals and ECG-derived respiration (EDR) from R waves of QRS amplitudes were used as inputs to the SVMs to recognize OSAS +/- subjects. Using leave-one-out technique, the maximum accuracy of classification for 83 training sets was found to be 100% for SVMs using a subset of selected combination of HRV and EDR features. Independent test results on 42 subjects showed that it correctly recognized 24 out of 26 OSAS + subjects and 15 out of 16 OSAS - subjects (accuracy = 92.85%; Cohen's kappa value of 0.85). For estimating the relative severity of OSAS, the posterior probabilities of SVM outputs were calculated and compared with respective apnea/hypopnea index. These results suggest superior performance of SVMs in OSAS recognition supported by wavelet-based features of ECG. The results demonstrate considerable potential in applying SVMs in an ECG-based screening device that can aid a sleep specialist in the initial assessment of patients with suspected OSAS.     

Variability, regularity, and complexity of time series generated by schizophrenic patients and control subjects

We analyzed time series generated by 20 schizophrenic patients and 20 sex- and age-matched control subjects using three nonlinear methods of time series analysis as test statistics: central tendency measure (CTM) from the scatter plots of first differences of data, approximate entropy (ApEn), and Lempel-Ziv (LZ) complexity. We divided our data into a training set (10 patients and 10 control subjects) and a test set (10 patients and 10 control subjects). The training set was used for algorithm development and optimum threshold selection. Each method was assessed prospectively using the test dataset. We obtained 80% sensitivity and 90% specificity with LZ complexity, 90% sensitivity, and 60% specificity with ApEn, and 70% sensitivity and 70% specificity with CTM. Our results indicate that there exist differences in the ability to generate random time series between schizophrenic subjects and controls, as estimated by the CTM, ApEn, and LZ. This finding agrees with most previous results showing that schizophrenic patients are characterized by less complex neurobehavioral and neuropsychologic measurements.     

Assessment of closed-loop ventilatory stability in obstructive sleep apnea

Previous studies on ventilatory control in obstructive sleep apnea (OSA) have generally indicated depressed chemosensitivity, implying greater stability of the chemical control of breathing in these subjects. However, these results were based on tests involving steady-state or quasi-steady measurements obtained in wakefulness. We have developed a method for assessing the dynamic stability characteristics of chemoreflex control in OSA patients during sleep. While continuous positive airway pressure was applied to stabilize the upper airways, acoustically stimulated arousals were used to perturb the respiratory system during sleep. The fluctuations in esophageal pressure that ensued were analyzed, using a closed-loop minimal model, to estimate the chemoreflex loop impulse response (CLIR). Tests using simulated data confirmed the validity of our estimation algorithm. Application of the method to arousal responses measured in six OSA and five normal subjects revealed no statistically significant differences in gain margins and loop gain magnitudes between the two groups. However, the CLIR in the OSA subjects exhibited faster and more oscillatory dynamics. This result implies that, in addition to unstable upper airway mechanics, an underdamped chemoreflex control system may be another important factor that promotes the occurrence of periodic obstructive apneas during sleep.     

A noninvasive technique for detecting obstructive and central sleepapnea

A new noninvasive method to detect obstructive and central sleep apnea [(OSA) and (CSA)] events is described. Data were collected from ten volunteer subjects with a previous diagnosis of OSA while they were titrated for continuous positive airway pressure (CPAP) therapy. Apneic events were identify by analyzing of estimated airway impedance determined from pressure and airflow signals delivered from CPAP. To enhance performance of this technique, a single-frequency (5 Hz with 0.5 cmH₂O peak-to-peak amplitude) probing signal was superimposed on the applied CPAP pressure. The results indicated that estimated airway impedance during OSA (mean: 17.9, SD: 3.4, N=50) was significantly higher then during CSA (mean: 4.1, SD: 1.7, N=50). When the estimated impedance of OSA and CSA events were compared to a fixed threshold, 100% of all events can be correctly categorized. These results indicate that it may be possible to diagnose OSA and CSA noninvasively based upon this technique. The instrument and the algorithm required are relatively simple and can be incorporated in a home-based device. If this method was used for prescreening apnea patients, it could reduce cost, waiting time, and discomfort associated with traditional diagnostic procedures     

An adaptive detector of genioglossus EMG reflex using Berkner transform for time latency measurement in OSA pathophysiological studies

To investigate obstructive sleep apnea syndrome mechanisms, we developed a device to measure the surface electromyogram (EMG) time latency reflex of the genioglossus muscle stimulated by time and amplitude calibrated negative pharyngeal pressure drops. The reflex signals were found to be disturbed by transient signals that generate false alarms. Thus, to reduce false alarm occurrences we designed an adaptive multiscale method. Continuous wavelet transform (CWT) is widely used in biomedical signal event detection processes. The Berkner transform is an approximation of a CWT that is based on a hierarchical scheme similar to discrete wavelet transform. We used the Berkner transform to build a multiscale detector because it offers the possibility of maxima coefficients linkage that leads to good accuracy in reflex onset localization. As a contribution to this novel approach we used a reconstruction formula to develop an adaptive method for scale range determination in our surface EMG reflex detector. Finally, we characterized our detector in terms of accuracy and robustness, first on synthesized signals and second, on signals acquired on apneic patients and healthy subjects. Preliminary results showed a significant difference (p < 0.01) between the two populations regarding the genioglossus muscle mean latency time. These physiological findings may partly explain why the upper airway protective reflex occurring when a negative pressure is applied to the upper airway is ineffective in OSA patients, leading to pharyngeal collapse.     

Snore signal enhancement and activity detection via translation-invariant wavelet transform

Acoustical properties of snores have been widely studied as a potentially cost-effective and reliable alternative to diagnosing obstructive sleep apnea (OSA), with a common recognition that the diagnostic accuracy depends heavily on the snore signal quality and intelligibility. This paper proposes a novel preprocessing system that performs two critical tasks concurrently in a translation-invariant wavelet transform domain. These tasks include enhancement of snore signals via a level-correlation-dependent (LCD) threshold, and identification of snore presence through a snore activity (SA) detector. Various experiments were conducted to warrant the robustness of the system in terms of theoretical statistics quality, signal-to-noise ratio, mean opinion score, and clinical usefulness in detecting OSA. Results indicate that the proposed LCD threshold and SA detector are highly comparable to the existing denoising methodologies using level-dependent threshold and segmentation approaches using short-time energy and zero-crossing rate, yielding the best results in all the experiments. Given the strong initial performance of the proposed preprocessing system for snore signals, continued exploration in this direction could potentially lead to additional improvement in signal integrity, thereby increasing the diagnostic accuracy for OSA.     

Microscopic image analysis for quantitative measurement and featureidentification of normal and cancerous colonic mucosa

The development of an automated algorithm for the categorization of normal and cancerous colon mucosa is reported. Six features based on texture analysis were studied. They were derived using the co-occurrence matrix and were angular second moment, entropy, contrast, inverse difference moment, dissimilarity, and correlation. Optical density was also studied. Forty-four normal images and 58 cancerous images from sections of the colon were analyzed. These two groups were split equally into two subgroups: one set was used for supervised training and the other to test the classification algorithm. A stepwise selection procedure showed that correlation and entropy were the features that discriminated most strongly between normal and cancerous tissue (P<0.0001). A parametric linear-discriminate function was used to determine the classification rule. For the training set, a sensitivity and specificity of 93.1% and 81.8%, respectively, were achieved, with an overall accuracy of 88.2%. These results mere confirmed with the test set, with a sensitivity and specificity of 93.1% and 86.4%, respectively, and an overall accuracy of 90.2%     

Evolutionary Maximum Entropy Spectral Estimation and Heart Rate Variability Analysis

Spectral analysis has been used extensively in heart rate variability (HRV) studies. The spectral content of HRV signals is useful in assessing the status of the autonomic nervous system. Although most of the HRV studies assume stationarity, the statistics of HRV signals change with time due to transients caused by physiological phenomena. Therefore, the use of time-frequency analysis to estimate the time-dependent spectrum of these non-stationary signals is of great importance. Recently, the spectrogram, the Wigner distribution, and the evolutionary periodogram have been used to analyze HRV signals. In this paper, we propose the application of the evolutionary maximum entropy (EME) spectral analysis to HRV signals. The EME spectral analysis is based on the maximum entropy method for stationary processes and the evolutionary spectral theory. It consists in finding an EME spectrum that matches the Fourier coefficients of the evolutionary spectrum. The spectral parameters are efficiently calculated by means of the Levinson algorithm. The EME spectral estimator provides very good time-frequency resolution, sidelobe reduction and parametric modeling of the evolutionary spectrum. With the help of real HRV signals we show the superior performance of the EME over the earlier methods.     

Anatomical optical coherence tomography for long-term, portable, quantitative endoscopy

In this paper, we report on anatomical optical coherence tomography, a catheter-based optical modality designed to provide quantitative sectional images of internal hollow organ anatomy over extended observational periods. We consider the design and performance of an instrument and its initial intended application in the human upper airway for the characterization of obstructive sleep apnea (OSA). Compared with current modalities, the technique uniquely combines quantitative imaging, bedside operation, and safety for use over extended periods of time with no cumulative dose limit. Our experiments show that the instrument is capable of imaging subjects during sleep, and that it can record dynamic changes in airway size and shape.     

睡眠脑电时间序列的非线性样本熵研究

比较了样本熵与近似熵算法的区别,通过对构造的一个由随机信号和确定性信号组成的混合系统进行分析可以看出在公差阈值小于0.2时,样本熵比近似熵更适合于时间序列信号的复杂度分析.然后,对采集的整夜睡眠脑电信号,用样本熵作为睡眠脑电数据的特征值,分析了睡眠过程不同阶段的实验数据.结果表明,不同睡眠时期样本熵有差别,随睡眠深度的加深,样本熵值变小.因此,样本熵可以很好地区分不同睡眠时期并作为睡眠自动分期的一个重要的非线性特征参数.     

QT variability and HRV interactions in ECG: quantification and reliability

In this paper, a dynamic linear approach was used over QT and RR series measured by an automatic delineator, to explore the interactions between QT interval variability (QTV) and heart rate variability (HRV). A low-order linear autoregressive model allowed to separate and quantify the QTV fractions correlated and not correlated with HRV, estimating their power spectral density measures. Simulated series and artificial ECG signals were used to assess the performance of the methods, considering a respiratory-like electrical axis rotation effect and noise contamination with a signal-to-noise ratio (SNR) from 30 to 10 dB. The errors found in the estimation of the QTV fraction related to HRV showed a nonrelevant performance decrease from automatic delineation. The joint performance of delineation plus variability analysis achieved less than 20% error in over 75% of cases for records presenting SNRs higher than 15 dB and QT standard deviation higher than 10 ms. The methods were also applied to real ECG records from healthy subjects where it was found a relevant QTV fraction not correlated with HRV (over 40% in 19 out of 23 segments analyzed), indicating that an important part of QTV is not linearly driven by HRV and may contain complementary information.     

Renyi entropy measures of heart rate Gaussianity

Sample entropy and approximate entropy are measures that have been successfully utilized to study the deterministic dynamics of heart rate (HR). A complementary stochastic point of view and a heuristic argument using the Central Limit Theorem suggests that the Gaussianity of HR is a complementary measure of the physiological complexity of the underlying signal transduction processes. Renyi entropy (or q-entropy) is a widely used measure of Gaussianity in many applications. Particularly important members of this family are differential (or Shannon) entropy (q = 1) and quadratic entropy (q = 2). We introduce the concepts of differential and conditional Renyi entropy rate and, in conjunction with Burg's theorem, develop a measure of the Gaussianity of a linear random process. Robust algorithms for estimating these quantities are presented along with estimates of their standard errors.