Mohammad A. Al-Abed

A Method to Detect Obstructive Sleep Apnea Using Neural Network Classification of Time-Frequency Plots of the Heart Rate Variability

This paper presents a new method of analyzing time-frequency plots of heart rate variability to detect sleep disordered breathing from nocturnal ECG. Data is collected from 12 normal subjects (7 males, 5 females; age 46 plusmn 9.38 years, AHI 3.75 plusmn 3.11) and 14 apneic subjects (8 males, 6 females; age 50.28 plusmn 9.60 years; AHI 31.21 plusmn 23.89). The proposed algorithm uses textural features extracted from normalized gray-level co-occurrence matrices (NGLCM) of images generated by short-time discrete Fourier transform (STDFT) of the HRV. Using feature selection, seventeen features extracted from 10 different NGLCMs representing four characteristically different gray-level images are used as inputs to a three-layer Multi-Layer Perceptron (MLP) classifier. After a 1000 randomized Monte-Carlo simulations, the mean training classification sensitivity, specificity and accuracy are 99.00%, 93.42%, and 96.42%, respectively. The mean testing classification sensitivity, specificity and accuracy are 94.42%, 85.40%, and 90.16%, respectively.

A New Method to Detect Obstructive Sleep Apnea Using Fuzzy Classification of Time-Frequency Plots of the Heart Rate Variability

This paper presents a new method of analyzing time frequency plots of heart rate variability to detect sleep disordered breathing from nocturnal ECG. Data is collected from 12 normal subjects (7 males, 5 females; age 46 +/- 9.38 years, AHI 3.75 +/- 3.11) and 14 apneic subjects (8 males, 6 females; age 50.28 +/- 9.60 years; AHI 31.21 +/- 23.89). The proposed algorithm uses textural features extracted from normalized gray-level co-occurrence matrices (NGLCM) of images generated by short-time discrete Fourier transform (STDFT) of the HRV. Thirty selected features extracted from 10 different NGLCMs representing four characteristically different gray-level images are used as inputs to 10 Fuzzy Logic Systems (FLS) Classifiers. Each FLS is trained and their outputs are combined using a weighed majority rule method. The mean training detection sensitivity, specificity and accuracy are 86.87%, 71.72%, and 79.29%, respectively. The mean testing detection sensitivity, specificity and accuracy are 83.22%, 68.54%, and 75.88%, respectively.

An investigation of simultaneous variations in cerebral blood flow velocity and arterial blood pressure during Sleep Apnea

Obstructive Sleep Apnea (OSA) is a major sleep disorder with a prevalence of about 15 % among US adult population and can lead to cardiovascular diseases and stroke. In this study, we have investigated the OSA-induced concurrent rise in cerebral blood flow velocity and blood pressure in 5 positively diagnosed sleep apnea subjects. The subject population had a mean AHI of 57.94±25.73 and BMI of 33.66±7.27 kg/m2. The results of this preliminary study yielded a relatively high correlation between rise in blood pressure and rise in cerebral blood flow velocity during apnea episodes (r=0.61±0.16) compared to normal breathing (r=0.28±0.26). These findings suggest that cerebral autoregulation may be less effective during apnea episodes.

Detection of airway occlusion in simulated obstructive sleep apnea/hypopnea using ultrasound: An in vitro study

Obstructive sleep apnea/hypopnea Syndrome (OSAHS) is the most common form of Sleep Disordered Breathing (SDB) and it is estimated to affect approximately 6% of US adult population. Various methods have been proposed for the development of inexpensive screening methods to detect SDB to reduce the need for costly nocturnal polysomnography (NPSG). By using the existing air in the airway as an ultrasonic contrast agent, we propose a method to examine the narrowing or occlusion of the airway associated with OSAHS events. We describe here an in vitro study that approximates the anatomical and acoustic characteristics of the airway and neck. In this experiment, we simulate the fully open airway as well as apnea and hypopnea events. These in vitro studies results show significant differences in the ultrasonic signals acquired from the open airway model versus those from the model depicting apnea and hypopnea events. Therefore, the findings provide a foundation for development of an ultrasound system to detect SDB in vivo.

Concurrent variations of cerebral blood flow and arterial blood pressure in simulated sleep apnea

Obstructive Sleep Apnea (OSA) is one of the most common sleep disordered breathing which affects about 15 % of US adult population. OSA is considered to be an important risk factor for the development of cardiac dysfunction and stroke. In this paper, we present the initial results of our investigation of the relationship between arterial blood pressure and cerebral blood flow velocity in simulated apnea. Sixteen healthy subjects (9 male, 7 female) of 29±4.89 yrs age and body mass index of 24.07±4.84 kg/m2 participated in the study. Our findings indicate that cerebral blood flow velocity variations has a relatively high correlation to changes in blood pressure during simulated apnea (r=0.74 ±0.06), suggesting that cerebral autoregulation may not compensate for the pressure changes during apnea.

Sleep disordered breathing detection using heart rate variability and R-peak envelope spectrogram

We report that combining the interbeat heart rate as measured by the RR interval (RR) and R-peak envelope (RPE) derived from R-peak of ECG waveform may significantly improve the detection of sleep disordered breathing (SDB) from single lead ECG recording. The method uses textural features extracted from normalized gray-level cooccurrence matrices of the time frequency plots of HRV or RPE sequences. An optimum subset of textural features is selected for classification of the records. A multi-layer perceptron (MLP) serves as a classifier. To evaluate the performance of the proposed method, single Lead ECG recordings from 7 normal subjects and 7 obstructive sleep apnea patients were used. With 500 randomized Monte-Carlo simulations, the average training sensitivity, specificity and accuracy were 100.0%, 99.9%, and 99.9%, respectively. The mean testing sensitivity, specificity and accuracy were 99.0%, 96.7%, and 97.8%, respectively.

In vivo characterization of ultrasonic transducers for the detection of airway occlusion in Sleep Disordered Breathing

Obstructive sleep apnea/hypopnea Syndrome (OSAHS) is the most common form of Sleep Disordered Breathing (SDB) and it is estimated to affect approximately 15% of US adult population. Various methods have been proposed for the development of inexpensive screening methods to detect SDB to reduce the need for costly nocturnal polysomnography (NPSG). In this paper, a description of the ultrasonic transducer design and characterization is presented, followed by the results of a full night sleep study. The findings show a significant difference in the temporal features extracted from the received ultrasonic waveform during apneic breathing, compared to the hyperventilation that follows. Therefore, the findings indicate the feasibility of developing an ultrasonic detection device for low cost diagnosis of SDB.

Upper airway occlusion detection using a novel ultrasound technique

Obstructive sleep apnea/hypopnea Syndrome (OSAHS) is the most common form of Sleep Disordered Breathing (SDB) and it is estimated to affect approximately 15% of US adult population. In this paper, we report on the results of in vivo experiments of an ultrasonic device for the non-invasive detection of obstructive sleep apnea/hypopnea (OSAH). A description of the ultrasonic system used is presented, followed by the results of a full night sleep study. The findings show a significant difference in the spectral features extracted from the received ultrasonic waveform during apneic breathing, compared to the hyperventilation that follows. Therefore, the findings indicate the feasibility of developing an ultrasonic detection device for low cost diagnosis of SDB.

Cross correlation and scatter plots of the heart rate variability and R-peak Envelope as features in the detection of obstructive sleep apnea

Interbeat heart rate as measured by the RR interval (RR) and R-Peak Envelope (RPE) are two signals that can be extracted from an Electrocardiogram (ECG) with relative ease and high reliability. RR and RPE have been shown to carry markers for detecting sleep disordered breathing (SDB). In this pilot study, we explore the cross correlation of RR and RPE in normal and SDB patients. Nocturnal ECG from 7 normal subjects and 7 SDB patients were used to obtain RR and RPE. The results revealed that the cross correlation of RR and RPE signals is significantly different between normal subjects and SDB patients (p < 2×10−6). Furthermore, a new scatter plot of RR vs. RPE was developed. Optimum features from the RR vs. RPE scatter plot were extracted and used as input to a multilayer perceptron (MLP) classifier to distinguish between normal and SDB subjects, The detection sensitivity, specificity and accuracy for the training data set were 95.0%, 100.0%, and 97.5%, respectively; and for the test data were 76.6%, 93.2%, and 84.7%, respectively.

Relation between arterial blood pressure and cerebral blood flow velocity in simulated sleep apnea

Obstructive Sleep Apnea (OSA) is one of the most common breathing disorder, affecting approximately 27% of U.S. adults. Limited data have suggested that OSA causes cerebral autoregulation impairment, thus being an important risk factor to stroke. The objective of this paper is to investigate and measure the relation between arterial blood pressure (BP) and cerebral blood flow velocity (CBFV) in simulated apnea. Sixteen healthy subjects (9 male, 7 female) of 29±4.89 yrs age and body mass index of 24.07±4.84 kg/m2 participated in the study. Four protocols were used; sitting 30 seconds, 90s, and supine 30s and 90s. Our results showed that systolic BP and peak CBFV were correlated with average r=0.672 +0.265. Also, CBFV exhibited a significantly higher percent rise than BP. Thus, our findings suggest that cerebral autoregulation may be impaired during apnea episodes.