ArticlePDF Available

Abstract and Figures

Obstructive sleep apnea (OSA) is a common disorder in which individuals stop breathing during their sleep. These episodes last 10 seconds or more and cause oxygen levels in the blood to drop. Most of sleep apnea cases are currently undiagnosed because of expenses and practicality limitations of overnight polysomnography (PSG) at sleep labs, where an expert human observer is required. New techniques for sleep apnea classification are being developed by bioengineers for most comfortable and timely detection. In this study, we develop and validate a neural network (NN) using SpO2 measurements obtained from pulse oximetry to predict OSA. The results show that the NN is useful as a predictive tool for OSA with a high performance and improved accuracy, approximately 93.3%, which is better than reported techniques in the literature.
Content may be subject to copyright.
(IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 3, No.5, 2012
7 | P a g e
www.ijacsa.thesai.org
A Neural Network System for Detection of
Obstructive Sleep Apnea Through SpO2 Signal
Features
Laiali Almazaydeh, Miad Faezipour, Khaled Elleithy
Department of Computer Science and Engineering
University of Bridgeport
Bridgeport, CT 06604, USA
AbstractObstructive sleep apnea (OSA) is a common disorder
in which individuals stop breathing during their sleep. These
episodes last 10 seconds or more and cause oxygen levels in the
blood to drop. Most of sleep apnea cases are currently
undiagnosed because of expenses and practicality limitations of
overnight polysomnography (PSG) at sleep labs, where an expert
human observer is required. New techniques for sleep apnea
classification are being developed by bioengineers for most
comfortable and timely detection. In this study, we develop and
validate a neural network (NN) using SpO2 measurements
obtained from pulse oximetry to predict OSA. The results show
that the NN is useful as a predictive tool for OSA with a high
performance and improved accuracy, approximately 93.3%,
which is better than reported techniques in the literature.
Keywords- sleep apnea; PSG; SpO2; features extraction; oximetry;
neural networks.
I. INTRODUCTION
A. Background
Excessive daytime sleepiness and fatigue are the most
symptoms of sleep disorders. The risk of sleepiness and fatigue
lead to poor judgment and reaction time, especially for the
drivers who do not take sleepiness seriously.
Sleep apnea is becoming a more common cause of
sleepiness in children and adults. It is characterized by
abnormal pauses of breathing or abnormally low breath during
sleep. These pauses of breathing can range in frequency and
duration. The duration of the pause might be ten to thirty
seconds and upwards to as much as four hundred per night in
those with severe sleep apnea [1].
Sleep apnea is classified into two types. The first type is
Obstructive Sleep Apnea (OSA), which is generally caused by
a collapse of the upper respiratory airway. The second one is
Central Sleep Apnea (CSA), which is caused by inhibited
respiratory drive, since the brain fails to appropriately control
breathing during sleep. Out of the two sleep apnea types, OSA
is more common than CSA [2].
Sleep apnea is not a problem to be taken lightly, since it is
associated with, and is the possible cause of other medical
conditions such as high blood pressure, heart disease, diabetes
and depression.
Statistics show that over 18 million Americans suffer from
sleep apnea, while an estimate of 10 million Americans remain
undiagnosed [3]. Most cases go undiagnosed because of the
inconvenience, expenses and unavailability of testing. Testing
is inconvenient to the patient because it requires them to spend
the night away from their bed, causing discomfort. It is
expensive because testing is done in the hospital, causing
machines and various technicians and staff to work over night.
Testing is also widely unavailable due to sleep centers
operating at full capacity, and those on the waiting list can be
untreated for an additional 6 months.
Polysomnography (PSG) is a test commonly ordered for
some sleep disorders. It records the breath airflow, respiratory
movement, oxygen saturation, body position,
electroencephalogram (EEG), electrooculogram (EOG),
electromyogram (EMG), and electrocardiogram (ECG) [4].
To summarize, the mere dependency on PSG needs to be
taken away from the laboratory for simpler detection and faster
treatment. In this regard, we present a work based on a neural
network using SpO2 features extraction that will be used in a
larger real time system for sleep apnea diagnosis. The objective
of the system is to alert a patient who might be subject to an
apnea attack.
B. Paper Organization
The rest of this paper is organized as follows. In Section II
we glance at a variety of sleep apnea detection methods.
Section III contains an overview of the system, including a
description of the database of subjects, and the details of the
analysis methodology including features extraction of the SpO2
signal. The Neural Networks we used in this work is also
described in the same Section. In Section IV, we detail the
results of our system. Finally, Section V concludes this paper
regarding the potential usefulness of our system, and highlights
some directions for future research.
II. RELATED WORK
Several methods have been suggested for identification of
sleep apnea over the past few years. Statistical features of
different signals such as nasal air flow, the thorax and abdomen
(IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 3, No.5, 2012
8 | P a g e
www.ijacsa.thesai.org
effort signals, acoustic speech signal, oxygen saturation,
electrical activity of the brain (EEG), and electrical activity of
the heart (ECG) are commonly used in the detection.
Quiceno-Manrique et al. [5] proposed a simple diagnostic
tool for OSA with a high accuracy (up to 92.67%) using time-
frequency distributions and dynamic features in ECG signal.
The relationship between periodic changes in the oxygen
saturation (SaO2) profile and in the EEG pattern due to apnea
events during the night was investigated in [6]. The spectral
analysis of these two signals achieved 91% sensitivity, 83.3%
specificity and 88.5% accuracy in OSA diagnosis.
Ng et al. [7] showed that thoracic and the abdominal signals
were good parameters for the identification of the occurrence
of sleep apnea. Using the mean of absolute amplitudes of the
thoracic and the abdominal signals, the authors achieved good
performance with a receiver operating characteristic value
higher than 80%.
Wavelet transforms and an artificial neural network
(ANN) algorithm were applied to the EEG signal in [8] to find
a solution to the problem of identifying sleep apnea (SA)
episodes. The system's identification results achieved a
sensitivity of approximately 69.64% and a specificity of
approximately 44.44%.
Based on spectral components of heart rate variability
(HRV), frequency analysis was performed in [9] to detect sleep
apnea. Using Fourier and Wavelet Transformation with
appropriate application of the Hilbert Transform, the sensitivity
was 90.8%. In addition, in [10] a bivariate autoregressive
model was used to evaluate beat-by-beat power spectral density
of HRV and R peak area, where the sleep apnea classification
results showed accuracy higher than 85%.
The study in [11] assesses the analysis of various feature
sets and a combination of classifiers based on the arterial
oxygen saturation signal measured by pulse oximetry (SpO2)
and the ECG in order to evaluate sleep quality. With selected
features of the SpO2 and ECG signals, the Bagging with REP
Tree classifier achieved sensitivity of 79.75%, specificity of
85.89% and overall accuracy of 84.40%.
III. METHODOLOGY
A. Subjects
The database of SpO2 signals used in this research is
available from the PhysioNet web site [12].
PhysioNet contains a growing collection of biomedical
signals from healthy subjects and patients. The PhysioNet web
site is a public service of the PhysioNet Resource funded by the
National Institutes of Health’s NIBIB and NIGMS. PhysioNet
offers free access to Apnea-ECG Database, which we use to
assess and validate our approach.
The Apnea-ECG Database contains 8 recordings with SpO2
signals. These recordings have varying length from slightly less
than 7 hours to nearly 10 hours each.
The subjects of these recordings were men and women
between 27 and 63 years of age (mean: 43.8±10.8 years) with
weights between 53 and 135 kg (mean: 86.3±22.2 kg). The
sleep recordings originated from 32 subjects (25 men, 7
female), who were recruited for previous studies on healthy
volunteers and patients with obstructive sleep apnea [4].
B. SpO2 Signal
SpO2 is the amount of oxygen being carried by the red
blood cell in the blood. Very simply, SpO2 goes up and down
Figure 1. SpO2 record with OSA negative subject [12].
Figure 2. SpO2 record with OSA positive subject [12].
according to how well a person is breathing and how well
the blood is being pumped around the body [13].
SpO2 measured by pulse oximetry can be useful in OSA
diagnosis. Significant changes can be found in patients affected
by OSA because of the recurrent episodes of apnea, which are
frequently accompanied by oxygen desaturations [14].
Figure 1 depicts a common OSA negative subject, and
Figure 2 shows a SpO2 record with OSA positive subject.
However, diagnosis of the disease is not evident by visual
inspection.
C. Features Extraction
In our work, the SpO2 signals are saved to separate files and
processed off-line by an automated system we developed using
MATLAB to compute two of the common oximetric indices
and one nonlinear metric.
These three features are detailed as follows, respectively:
1) Delta index (∆ index): This is a common measure to
detect the apneic events by measuring SpO2 variability. Levy et
al. [15] calculates ∆ index as the sum of the absolute variations
between two successive points, divided by the number of
intervals. It is usually computed for 12-sec. intervals.
2) Oxygen desaturation indices of 3% (ODI3): This
measure is obtained by calculating the number of times per
hour with values of SpO2 greater than or equal to 3% from the
baseline. The baseline is set initially as the mean level in the
first 3 minutes of recording [16].
(IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 3, No.5, 2012
9 | P a g e
www.ijacsa.thesai.org
Central tendency measure with radius 0.5 (CTM50): This
measure applied in [16]. CTM50 is computed by selecting a
circular region of radius 0.5 around the origin, counting the
number of points that fall within the radius, and dividing by the
total number of points. Given N data points from a time series,
N-2 would be the total number of points in the scatter plot.
Hence, CTM50 can be computed as [17]:

2
)(
2
1
N
d
CTM
i
N
i

where,
 
 
5.0 1 1/2
22 112 iiii xxxxif
)( id
(2)
otherwise 0
D. Multi layer Networks Classifiers
In this research, we apply a neural network (NN) as a
classifier to identify the diagnostic performance of OSA using
SpO2 features.
A neural network is used to perform a pattern classification
task. NNs classifiers have been proven to be extremely helpful
in assisting medical specialties in clinical diagnosis [18].
The NN described in this study is based on three layers
feed-forward neural network learned with back-propagation
algorithm; an input layer, an output layer, and a hidden layer.
The hidden layer consists of a direct connection between the
input and the output layer.
The three SpO2 features act as inputs to a neural network,
and the diagnosis of OSA is the target. This is achieved by
presenting previously recorded inputs to a neural network and
then tuning it to produce the desired target outputs. This
process is called neural network training.
A total of 93data sets (41 with a positive diagnosis of OSA
and 52 with a negative diagnosis of OSA) are used. Validation
is done with the same training dataset, and test dataset has been
set to 17% of the original data. With these settings, the input
vectors and target vectors will be randomly divided into two
sets as follows:
83% are used for training and validation.
The last 17% are used as a completely independent test of
network generalization.
The training set with 78 samples was used to train the
network. Network parameters are adjusted through training by
attempting to minimize the error between the target (t) and the
actual (y) network output values. This error is expressed as the
mean square error [19]:
2
1
1
N
n
nn yt
N
E
(3)
where N is the number of samples in the training set.
In the training phase, the Purelin linear transfer function has
been used as an activation function of the output layer of the
network (for improving error derivative) [18]. Since the output
space must be divided into two regions: OSA positive and
OSA negative, we suggest using a single output node.
Figure 3. Confusion matrix for training set classification.
Figure 4. Confusion matrix for testing set classification.
We applied Hardlim function [18] to test the data to improve
the output of the network in the validation and testing phases.
IV. RESULTS
A. Performance Evaluation
We evaluated the classification performance of the selected
network configurations on the test set. Sensitivity, specificity
for testing data and accuracy are computed. A confusion matrix
is generated for the NN evaluation.
(IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 3, No.5, 2012
10 | P a g e
www.ijacsa.thesai.org
Figures 3 and 4 show the confusion matrix for training and
testing set classification, respectively. The confusion matrix
shows the total percent of correctly classified cases and the
total percent of misclassified cases.
The results show very good network validation
performance of 85% and high diagnostic performance with an
accuracy of 93.3% correct detection rate (sensitivity 87.5%,
and specificity 100%).
As a result, to reduce the dependency on complex PSG test
measures, we find that NN using SpO2 measurements is a
practical and useful screening test to estimate whether patients
have sleep apnea or not.
TABLE I.
Comparison of Sleep Apnea Detection Approaches.
Ref.
Approach
Performance [%]
Se
Sp
Acc.
[5]
ECG signal
92.67
[6]
SaO2 and EEG
signal
91
83.3
88.5
[7]
Thoracic and
abdominal signals
NA
NA
80
[8]
EEG signal
69.64
44.44
NA
[9]
HRV Fourier and
Wavelet
Transformation
90.8
NA
NA
[10]
Bivariate
autoregressive
model of HRV
NA
NA
85
[11]
SpO2 and ECG
79.75
85.89
84.40
-
Features
extraction of SpO2
signal
87.5
100
93.3
B. Comparison With other Works
We performed a comparison with other sleep apnea
detection techniques. Table I represents comparative results. As
can be seen, our system has achieved a comparable or better
performance. This applies to the other works that rely on the
SpO2 signal as well as other biometric signals.
V. CONCLUSION AND FUTURE DIRECTIONS
In this work, we studied the possibility of the detection of
sleep apnea from the SpO2 signal variation patterns. We further
developed a NN using the SpO2 signal features and evaluated
its effectiveness. This study has demonstrated a high
performance and an improved accuracy of the NN.
A future direction to this work would be to apply our
methodology to a larger population to validate the results
obtained with this radius.
As another future direction, we are also planning to
compute the CTM with several radii for every SpO2 signal in
both OSA positive and OSA negative groups. Then, we select
the optimum radius that achieves the most significant
differences.
Moreover, we are planning to analyze the ECG features
signals, in order to use it along with the SpO2 signals to build a
two parameter technique and apply that as a system for
automated recognition of OSA.
ACKNOWLEDGMENT
We would like to express special thanks to Ahmad ElSayed
for his valuable suggestions in designing the Neural Network
for this study.
REFERENCES
[1] Sleep Disorders Guide. www.sleepdisordersguide.com.
[2] S. Isa, M. Fanany, W. Jatmiko and A. Arymurthy,“Sleep Apnea
Detection from ECG Signal, Analysis on Optimal Features, Principal
Components, and Nonlinearity,” in Proceedings of the 5th IEEE
International Conference on Bioinformatics and Biomedical
Engineering (iCBBE), pp. 1-4, May 2011.
[3] SleepMedInc. www.sleepmed.md.
[4] P. Chazal, T. Penzel and C. Heneghan, Automated Detection of
Obstructive Sleep Apnoea at Different Time Scales Using the
Electrocardiogram,” Institute of Physics Publishing, vol. 25, no. 4, pp.
967-983, Aug. 2004.
[5] A.F. Quiceno-Manrique, J.B. Alonso-Hernandez, C.M. Travieso-
Gonzalez, M.A. Ferrer-Ballester and G. Castellanos-Dominguez,
“Detection of Obstructive Sleep Apnea in ECG Recordings using Time-
Frequency Distributions and Dynamic Features,” in Proceedings of the
31st IEEE International Conference on Engineering in Medicine and
Biology Society (EMBS 2009), pp. 5559-5562, Sep. 2009.
[6] D. Avarez, R. Hornero, J. Marcos, F. Campo and M. Lopez, Spectral
Analysis of Electroencephalogram and Oximetric Signals in Obstructive
Sleep Apnea Diagnosis,” in Proceedings of the 31st IEEE International
Conference on Engineering in Medicine and Biology Society (EMBS
2009), pp. 400-403, Sep. 2009.
[7] A. Ng, J. Chung, M. Gohel, W. Yu, K. Fan and T. Wong, Evaluation of
the Performance of Using Mean Absolute Amplitude Analysis of
Thoracic and Abdominal Signals for Immediate Indication of Sleep
Apnoea Events,” Journal of Clinical Nursing, vol. 17, no. 17, pp. 2360-
2366, Sep. 2008.
[8] R. Lin, R. Lee, C. Tseng, H. Zhou, C. Chao, J. Jiang,“A Ne w Approach
for Identifying Sleep Apnea Syndrome Using Wavelet Transform and
Neural Networks,” Biomedical Engineering: Applications, Basis &
Communications, vol. 18, no. 3, pp. 138-143, 2006.
[9] M. Schrader, C. Zywietz, V. Einem, B. Widiger, G. Joseph, “Detection
of Sleep Apnea in Single Channel ECGs from the PhysioNet Data
Base,” Computers in Cardiology 2000, vol. 27, pp. 263-266, Sep. 2000.
[10] M. Mendez, D. Ruini, O. Villantieri, M. Matteucci, T. Penzel and A.
Bianchi, “Detection of Sleep Apnea from Surface ECG Based on
Features Extracted by an Autoregressive Model, in Proceedings of the
IEEE International Conference on Engineering in Medicine and Biology
Society (EMBS 2007), pp. 6105-6108, Aug. 2007.
[11] B. Xie, H. Minn, Real Time Sleep Apnea Detection by Classifier
Combination,” in IEEE Transactions on Information Technology in
Biomedicine (in Press), 2012.
[12] PhysioNet, www.physionet.org.
[13] What is SpO2. http://www.neann.com/spo2.htm.
[14] J. Marcos, R. Hornero, D. Alvarez, F. Campo, C. Zamrron and M.
Lopez, Single Layer Network Classifiers to Assist in the Detection of
Obstructive Sleep Apnea Syndrome from Oximetry Data,” in
Proceedings of the 30th IEEE International Conference on Engineering
in Medicine and Biology Society (EMBS 2008), pp. 1651-1654, Aug.
2008.
[15] P. Levy, J. Pepin, C. Blanc, B. Paramelle and C. Brambilla, Accuracy
of Oximetry for Detection of Respiratory Disturbances in Sleep Apnea
Syndrome,” Chestjournal, vol. 109, no. 2, pp. 395-399, Feb. 1996.
[16] D. Alvarez, R. Hornero, D. Abasolo, F. Campo and C. Zamarron,
Nonlinear Characteristics of Blood Oxygen Saturation from Nocturnal
Oximetry for Obstructive Sleep Apnoea Detection,” Institute of Physics
Publishing, vol. 27, no. 4, pp. 399-412, Apr. 2006.
(IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 3, No.5, 2012
11 | P a g e
www.ijacsa.thesai.org
[17] J. Jeong, J. Gore and B. Peterson, A Method for Determinism in Short
Time Series, and its Application to Stationary EEG,” in IEEE
Transactions on Biomedical Engineering, vol. 49, no. 11, pp. 1374-1379,
Nov. 2002.
[18] A. El-Solh, M. Mador, E. Brock, D. Shucard, M. Abdul-Khoudoud and
B. Grant, Validity of Neural Network in Sleep Apnea, Sleep journal,
vol. 22, no. 1, pp. 105-111, 1999.
[19] J. Marcos, R. Hornero, D. Alvarez, F. Campo and Miguel Lopez,
Applying Neural Network Classifiers in the Diagnosis of the
Obstructive Sleep Apnea Syndrome from Nocturnal Pulse Oximetric
Recordings,” in Proceedings of the 29th IEEE International Conference
on Engineering in Medicine and Biology Society (EMBS 2007), pp.
5174-5177, Aug. 2007.
... Various methods for sleep apnea detection based on deep learning were reviewed in [8]. In literature, most sleep-apnea detection algorithms using deep learning methods exhibit a resolution of 1 minute ie., inferences are obtained on a per-minute basis [6], [9], [10]. Moreover, SpO2 based sleep apnea detection algorithms are often accompanied by other signals to improve inferences [5], [6]. ...
... Mostafa et al. [9] Almazayadeh et al. [10] Xie et al. [6] Cen et al. [5] This work The performance of the proposed 1D-CNN networks is compared with that of state-of-the-art algorithms in Table III. It can be seen that Model 1 and Model 2 exhibit comparable performance in terms of accuracy but higher sensitivity to methods proposed in [9] which is tested on a combined database, and therefore a direct comparison to our method is not possible. ...
... It can be seen that Model 1 and Model 2 exhibit comparable performance in terms of accuracy but higher sensitivity to methods proposed in [9] which is tested on a combined database, and therefore a direct comparison to our method is not possible. Model 1 and Model 2 outperform the method proposed in [10] in terms of accuracy, which is tested on the Physionet Apnea Database. Model 1 and Model 2 outperforms the methods proposed in [6] and [5] in terms of accuracy and sensitivity. ...
Preprint
Full-text available
The abnormal pause or rate reduction in breathing is known as the sleep-apnea hypopnea syndrome and affects the quality of sleep of an individual. A novel method for the detection of sleep apnea events (pause in breathing) from peripheral oxygen saturation (SpO2) signals obtained from wearable devices is discussed in this paper. The paper details an apnea detection algorithm of a very high resolution on a per-second basis for which a 1-dimensional convolutional neural network -- which we termed SomnNET -- is developed. This network exhibits an accuracy of 97.08% and outperforms several lower resolution state-of-the-art apnea detection methods. The feasibility of model pruning and binarization to reduce the computational complexity is explored. The pruned network with 80% sparsity exhibited an accuracy of 89.75%, and the binarized network exhibited an accuracy of 68.22%. The performance of the proposed networks is compared against several state-of-the-art algorithms.
... Relying on the statistical and spectral features, [8] employed the use of bootstrap aggregating for the classification of a single -lead ECG, the method yielded 84.14% sensitivity, 86.83% specificity, and 85.97% accuracy. In [14], the extracted features of arterial oxygen saturation signal (Sp02) measured in PSG and the ECG are combined to identify apnea episodes. These features are used to train, test, and validate a neural network, a high accuracy of 93.3% was gotten. ...
Article
Full-text available
Sleep Apnea is an anomaly in sleeping characterized by short pause in breathing. Failure to treat sleep apnea leads to fatal complications in both psychological and physiological being of human. Electroencephalogram (EEG) performs an important task in probing for sleep apnea through identifying and recording the brain’s activities while sleeping. In this study, computer aided detection of sleep apnea from EEG signals is developed to optimize and increase the prompt recognition and diagnosis of sleep apnea in patients. The time domain, wavelets, and frequency domain of the EEG signals were computed, and features were extracted from these domains. These features are inputted into two machine learning algorithms: Support Vector Machine and K-Nearest Neighbors of different kernel functions and orders. Evaluation metrics such as specificity, accuracy, and sensitivity are computed and analyzed for the classifiers. The KNN classifier outperforms the SVM in classifying apnea from non-apnea events in patients. The KNN order 3 shows the highest performance sensitivity of 85.92%, specificity of 80% and accuracy of 82.69%.
... The highest EB sensitivity (97%) was reported by a method using oximetry analysis [165], while the highest SB sensitivity (100%) was attained by either oximetry [157] or ECG [179] [180] [195]. The maximum EB specificity (100%) was also attained using either oximetry [156] or ECG [176]. Regarding the best SB specificity (100%), the best results were achieved using either ECG [179] [193] [197] or sound [219]. ...
... Several approaches based on SpO2 data only have also been proposed in the literature [42,43,44,45,46,47,48,49]. Among them, Mostafa et al. [49] directly exploit the SpO2 signal following a deep learning approach, while the other ones extract some features from SpO2 data, that are then fed to a proper classification model. ...
Article
Obstructive Sleep Apnea Syndrome (OSAS) is the most common sleep-related breathing disorder. It is caused by an increased upper airway resistance during sleep, which determines episodes of partial or complete interruption of airflow. The detection and treatment of OSAS is particularly important in patients who suffered a stroke, because the presence of severe OSAS is associated with higher mortality, worse neurological deficits, worse functional outcome after rehabilitation, and a higher likelihood of uncontrolled hypertension. The gold standard test for diagnosing OSAS is polysomnography (PSG). Unfortunately, performing a PSG in an electrically hostile environment, like a stroke unit, on neurologically impaired patients is a difficult task; moreover, the number of strokes per day vastly outnumbers the availability of polysomnographs and dedicated healthcare professionals. Hence, a simple and automated recognition system to identify OSAS cases among acute stroke patients, relying on routinely recorded vital signs, is highly desirable. The vast majority of the work done so far focuses on data recorded in ideal conditions and highly selected patients, and thus it is hardly exploitable in real-life circumstances, where it would be of actual use. In this paper, we propose a novel convolutional deep learning architecture able to effectively reduce the temporal resolution of raw waveform data, like physiological signals, extracting key features that can be used for further processing. We exploit models based on such an architecture to detect OSAS events in stroke unit recordings obtained from the monitoring of unselected patients. Unlike existing approaches, annotations are performed at one-second granularity, allowing physicians to better interpret the model outcome. Results are considered to be satisfactory by the domain experts. Moreover, through tests run on a widely-used public OSAS dataset, we show that the proposed approach outperforms current state-of-the-art solutions.
... Relying on the statistical and spectral features, [8] employed the use of bootstrap aggregating for the classification of a single -lead ECG, the method yielded 84.14% sensitivity, 86.83% specificity, and 85.97% accuracy. In [14], the extracted features of arterial oxygen saturation signal (Sp02) measured in PSG and the ECG are combined to identify apnea episodes. These features are used to train, test, and validate a neural network, a high accuracy of 93.3% was gotten. ...
Article
Full-text available
Sleep Apnea is an anomaly in sleeping characterized by short pause in breathing. Failure to treat sleep apnea leads to fatal complications in both psychological and physiological being of human. Electroencephalogram (EEG) performs an important task in probing for sleep apnea through identifying and recording the brain’s activities while sleeping. In this study, computer aided detection of sleep apnea from EEG signals is developed to optimize and increase the prompt recognition and diagnosis of sleep apnea in patients. The time domain, wavelets, and frequency domain of the EEG signals were computed, and features were extracted from these domains. These features are inputted into two machine learning algorithms: Support Vector Machine and K-Nearest Neighbors of different kernel functions and orders. Evaluation metrics such as specificity, accuracy, and sensitivity are computed and analyzed for the classifiers. The KNN classifier outperforms the SVM in classifying apnea from non-apnea events in patients. The KNN order 3 shows the highest performance sensitivity of 85.92%, specificity of 80% and accuracy of 82.69%.
... Mostafa et al. [25] employed 61 time and frequency domain features along with a neural network (NN) and got maximum sensitivity of 96.5% with the SAE database. Similarly, Almazaydeh and Faezipour [1] used statical features, delta index, and a multi-layer NN for sleep apnea detection. They obtained an accuracy of 93.3% and sensitivity of 87.5% by taking 17% of the SAE database as test data. ...
Article
Full-text available
Sleep apnea is a potential sleep disorder, which deteriorates the quality of sleep. It is characterized by the obstruction in nasal airflow, which results in a low concentration of oxygen in the blood. Though polysomnography (PSG) is considered as a gold standard for diagnosing sleep apnea, it is arduous, demanding, expensive and inconvenient to the patients. This study presents an effective, efficient and sustainable sleep apnea automated detection system using pulse oximetry signals (SpO2), which indicate the percentage of oxygen content in the blood. The conventional methods, which employ PSG recordings are computationally intensive and costly. Nowadays, the focus is on non-invasive and portable devices for higher convenience and cost-effective diagnosis. In this work, we have used optimal duration-bandwidth concentrated wavelet transform to decompose the SpO2 signals into various subbands (SBs). The Shannon entropy features are extracted from various SBs coefficients. These features are then fed to various supervised machine learning algorithms, including decision trees and ensemble algorithms for automated detection of sleep apnea. The proposed model has attained the highest accuracy of 95.97% , and area under the receivers operating characteristics curve (AUC) of 0.98 for optimal wavelet-based Shannon entropy features when an ensemble boosting technique called random under-sampling boosting (RUSBoost) is employed with ten-fold cross-validation strategy. Thus, the proposed model is portable, economical, and accurate which can be used even at homes.
Conference Paper
Full-text available
Sleep apnea is a sleep disorder that can cause serious health problems. An Artificial Neural Network classifier to detect sleep apnea has been presented in this paper by utilizing the ECG signals. Moreover, the discrete wavelet transform is used to decompose the ECG signal and use the first decomposition for feature extraction; the extracted features were used to train the Artificial Neural Network for pattern detection using MATLAB tools. Also, the data-sets used contains both Apnea pat1ients and healthy volunteers' ECG signals. The results achieve 92.3% accuracy in the testing records.
Conference Paper
Full-text available
The abnormal pause or rate reduction in breathing is known as the sleep-apnea hypopnea syndrome and affects the quality of sleep of an individual. A novel method for the detection of sleep apnea events (pause in breathing) from peripheral oxygen saturation (SpO2) signals obtained from wearable devices is discussed in this paper. The paper details an apnea detection algorithm of a very high resolution on a per-second basis for which a 1-dimensional convolutional neural network- which we termed SomnNET- is developed. This network exhibits an accuracy of 97.08% and outperforms several lower resolution state-of-the-art apnea detection methods. The feasibility of model pruning and binarization to reduce the computational complexity is explored. The pruned network with 80% sparsity exhibited an accuracy of 89.75%, and the binarized network exhibited an accuracy of 68.22%. The performance of the proposed networks is compared against several state-of-the-art algorithms.
Article
Background and Objective Sleep Apnea Syndrome (SAS) is a multimorbid chronic disease with individual and societal deleterious consequences. Polysomnography (PSG) is the multi-parametric reference diagnostic tool that allows a manual quantification of the apnea-hypopnea index (AHI) to assess SAS severity. The burden of SAS is affecting nearly one billion people worldwide explaining that SAS remains largely under-diagnosed and undertreated. The development of an easy to use and automatic solution for early detection and screening of SAS is highly desirable. Methods We proposed an Accelerometry-Derived Respiratory index (ADR) solution based on a dual accelerometry system for airflow estimation included in a machine learning process. It calculated the AHI thanks to a RUSBoosted Tree model and used physiological and explanatory specifically developed features. The performances of this method were evaluated against a configuration using gold-standard PSG signals on a database of 28 subjects. Results The AHI estimation accuracy, specificity and sensitivity of the ADR index were 89%, 100% and 80% respectively. The added value of the specifically developed features was also demonstrated. Conclusion Overnight physiological monitoring with the proposed ADR solution using a machine learning approach provided a clinically relevant estimate of AHI for SAS screening. The physiological component of the solution has a real interest for improving performance and facilitating physician's adhesion to an automatic AHI estimation.
Article
Obstructive Sleep Apnea (OSA) is a potentially common sleep disorder in which the upper airways are collapsed either partially or completely. The golden standard method for treating OSA, is the full night Continuous Positive Airway Pressure (CPAP). Yet, due to the ensuing discomfort, it incurs on patients, researchers have been motivated to investigate other alternatives, whereby, OSA can be effectively treated. Recently, an increasingly popular OSA treatment has been developed that consists in activating the protrusion muscles of the tongue by stimulating the Hypoglossal Nerve (HGN). In this context, the present work is conducted to propose the design of apnea detector module as part of an implantable HGN stimulator based on the esophageal Pressure Pes signal as a new approach for controlling OSA occurrence. Specifically, an effective real-time apnea event detecting algorithm is put forward. Following the achievement of satisfactory simulation results, attained through the Modelsim simulation tool, we proceeded with assessing the possibility of its hardware implementation on a Field-Programmable Gate Array (FPGA) device. To this end, the apnea detector module was synthesized and designed. The low power consumption and the small size, characterizing this module, which have made it possible to integrate it as part of a wirelessly-powered implantable HGN stimulator.
Article
Full-text available
This paper describes implementation of Principal Component Analysis (PCA) on sleep apnea detection using Electrocardiogram (ECG) signal. The statistics of RR-intervals per epoch with 1 minute duration were used as an input. The combination of features proposed by Chazal and Yilmaz was transformed into orthogonal features using PCA. Cross validation, random sampling, and test on train data were used on model selection. The results of classification using kNN, Naïve- Bayes, and Support Vector Machine (SVM) show that PCA features give better classification accuracy compared to Chazal and Yilmaz features. SVM with RBF (Radial Basis Function) kernel gives the best classification accuracy by using 7 principal components (PC) as a features. The experimental results show that relation between Chazal features with target class tend to be linear, but Yilmaz and PCA features are non-linear.
Article
Full-text available
This paper describes a new technique to classify and analyze the electroencephalogram (EEG) signal and recognize the EEG signal characteristics of Sleep Apnea Syndrome (SAS) by using wavelet transforms and an artificial neural network (ANN). The EEG signals are separated into Delta, Theta, Alpha, and Beta spectral components by using multi-resolution wavelet transforms. These spectral components are applied to the inputs of the artificial neural network. We treated the wavelet coefficient as the kind of the training input of artificial neural network, might result in 6 groups of wavelet coefficients per second signal by way of characteristic part processing technique of the artificial neural network designed by our group, we carried out the task of training and recognition of SAS symptoms. Then the neural network was configured to give three outputs to signify the SAS situation of the patient. The recognition threshold for all test signals turned out to have a sensitivity level of approximately 69.64 and a specificity value of approximately 44.44 . In neurology clinics, this study offers a clinical reference value for identifying SAS, and could reduce diagnosis time and improve medical service efficiency.
Conference Paper
Full-text available
Detection of obstructive sleep apnea can be performed through heart rate variability analysis, since fluctuations of oxygen saturation in blood cause variations in the heart rate. Such variations in heart rate can be assessed by means of time-frequency analysis implemented with time-frequency distributions belonging to Cohen's class. In this work, dynamic features are extracted from time frequency distributions in order to detect obstructive sleep apnea from ECG signals recorded during sleep. Furthermore, it is applied a methodology to measure the relevance of each dynamic feature, before the implementation of k-nn classifier used to recognize the normal and pathologic signals. As a result, the proposed method can be applied as a simple diagnostic tool for OSA with a high accuracy (up to 92.67%) in one-minute intervals.
Article
Full-text available
A novel method for detecting determinism in short time series is developed and applied to investigate determinism in stationary electroencephalogram (EEG) recordings. This method is based on the observation that the trajectory of a time series generated from a differentiable dynamical system behaves smoothly in an embedded state space. The angles between two successive tangent vectors in the trajectory reconstructed from the time series is calculated as a function of time. The irregularity of the angle variations obtained from the time series is estimated using second-order difference plots, and compared with that of the corresponding surrogate data. Using this method, we demonstrate that scalp EEG recordings from normal subjects do not exhibit a low-dimensional deterministic structure. This method can be useful for analyzing determinism in short time series, such as those from physiological recordings.
Article
Full-text available
An automated classification algorithm is presented which processes short-duration epochs of surface electrocardiogram data derived from polysomnography studies, and determines whether an epoch is from a period of sleep disordered respiration (SDR) or normal respiration (NR). The epoch lengths considered were 15, 30, 45, 60, 75, and 90 s. Epochs were labeled as 'NR' or 'SDR' by a human expert, based on standard polysomnography interpretation rules. The automated classification algorithm was trained and tested on a database of 70 overnight ECG recordings from subjects with and without obstructive sleep apnoea syndrome (35 used for training, 35 for independent validation). Depending on the epoch length, the classifier correctly labeled between 87% (15 s epochs) and 91% (60 s epochs) of the epochs in the test set. Accuracy was lowest for the shortest (15 s) and longest (90 s) epoch lengths, but the analysis was relatively insensitive to choice of epoch length. The classifications from these epochs were combined to form an overall summary measure of minutes-of-SDR, allowing per-subject classification.
Article
To find an efficient and valid alternative of polysomnography (PSG), this paper investigates real-time sleep apnea and hypopnea syndrome (SAHS) detection based on electrocardiograph (ECG) and saturation of peripheral oxygen (SpO(2)) signals, individually and in combination. We include ten machine-learning algorithms in our classification experiment. It is shown that our proposed SpO (2) features outperform the ECG features in terms of diagnostic ability. More importantly, we propose classifier combination to further enhance the classification performance by harnessing the complementary information provided by individual classifiers. With our selected SpO(2) and ECG features, the classifier combination using AdaBoost with Decision Stump, Bagging with REPTree, and either kNN or Decision Table achieves sensitivity, specificity, and accuracy all around 82% for a minute-based real-time SAHS detection over 25 sleep-disordered-breathing suspects' full overnight recordings.
Article
This study assessed the hypothesis that blood oxygen saturation (SaO(2)) and electroencephalogram (EEG) recordings could provide complementary information in the diagnosis of the obstructive sleep apnea (OSA) syndrome. We studied 148 patients suspected of suffering from OSA. Classical spectral parameters based on the relative power in specified frequency bands (A(f-band)) or peak amplitudes (PA) were used to characterize the frequency content of SaO(2) and EEG recordings. Additionally, the median frequency (MF) and the spectral entropy (SE) were applied to obtain further spectral information. We applied a forward stepwise logistic regression (LR) procedure with crossvalidation leave-one-out to obtain the optimum spectral feature set. Two features from the oximetric spectral analysis (PA and MFsat) and three features from the EEG spectral analysis (A(delta), A(alpha) and SEeeg) were automatically selected. 91.0% sensitivity, 83.3% specificity and 88.5% accuracy were obtained. These results suggest that MF and SE could provide additional information to classical frequency characteristics commonly used in OSA diagnosis. Additionally, nocturnal SaO(2) and EEG recordings during the whole night could provide complementary information to help in the detection of OSA syndrome.
Article
The aim of this study is to assess the utility of single layer network classifiers to help in the diagnosis of the obstructive sleep apnea syndrome (SAOS). Oxygen saturation (SaO(2)) recordings from a total of 157 subjects suspected of suffering from OSAS were used. These were divided into a training set and a test set with 74 and 83 subjects, respectively. Four classification schemes were developed by using generalized linear models (GLM). Two GLM classifiers were built with spectral (GLM-SP) and non-linear (GLM-NL) features from SaO(2) signals, respectively. In addition, both algorithms were combined in order to improve their classification capability. The performance of two different ensemble classifiers was analyzed. The highest diagnostic accuracy was reached by the GLM-SP classifier (88%). The ensemble built from the combination of GLM-SP and GLM-NL by means of an additional GLM structure provided the best sensitivity value (87.8%). Applying spectral and non-linear features from SaO(2) data simultaneously could be useful in OSAS diagnosis.
Article
To evaluate the sensitivity of mean absolute amplitudes of the thoracic and the abdominal signals as a prompt indicator of the occurrence of sleep apnoea events. To provide symptomatic management of sleep apnoea, a reliable method of detecting sleep apnoea is essential to ensure that the intervention can be applied only when needed. It is also crucial to identify the threshold for the trigger of an intervention using a deployed sensor. Twenty-six subjects aged between 18-65 years who were diagnosed with obstructive or central sleep apnoea underwent an overnight sleep study. Signals of nasal and oral airflow, thoracic and abdominal efforts and pulse oximetry level were recorded using a polysomnography device. With a 95% CI, the overall area under the receiver operating characteristic of the thoracic signal, the abdominal signal and the combination of the thoracic and the abdominal signals were 84.56, 87.48 and 90.91%, respectively. Using -20, -25 and -30% as a cut-off point, the sensitivity values of thoracic signal, abdominal signal and combination of the thoracic and the abdominal signals ranged from 70.29-86.25% and the specificity values ranged from 74.82 to 90.09%. Using mean absolute amplitude analysis, the results of this study showed that combination of the thoracic and the abdominal signals achieved the best overall and individual performances compared with thoracic signal and abdominal signal. Overall, thoracic signal, abdominal signal and combination of the thoracic and the abdominal signals have a good performance with an receiver operating characteristic value higher than 80%. The thoracic and the abdominal signals were good parameters for the identification of the occurrence of sleep apnoea, being as quick as the nasal airflow signal. These results suggested that sleep apnoea events could be identified through constant monitoring of the patient's thoracic and abdominal signals. Knowledge of these signals could help nurses to manage sleep apnoea in patients.
Article
The cost and inconvenience of polysomnography make simplified techniques of screening desirable in the strategy of diagnosis of sleep apnea syndrome (SAS). We have evaluated, in a prospective study of 301 consecutive patients referred for suspected sleep disorders, an index (delta index) that detects apneic events by quantifying arterial oxygen saturation (SaO2) variability. Regional sleep laboratory taking referrals from general practitioners and specialists. Classic polysomnography was the gold standard, with 15 apneas plus hypopneas per hour (RDI) being used as a threshold for definition of obstructive sleep apnea (OSA). Oximetry was recorded over the same night. Signal variability was quantified as a function of time, using digital processing of oximetric data. Sensitivity, specificity, and positive and negative predictive values of oximetry testing were calculated. A receiver operating characteristic (ROC) curve was constructed representing the comparative courses of sensitivity and 1-specificity at different thresholds of delta index. Three hundred one patients were included (age, 56 +/- 12 years). Their RDI was 30 +/- 24. For a delta threshold at 0.6, the sensitivity of oximetry for the diagnosis of OSA was 98% and the specificity was 46%. The positive and negative predictive values for diagnosing SAS were 77% and 94%, respectively. The three false-negative cases had a relatively high awake SaO2 (97 vs 93.9 +/- 2.8%), a moderate RDI (23.3 +/- 1.6), and were less obese than the other patients (body mass index: 25 +/- 3 vs 33 +/- 8). The 58 false-positive cases had an RDI of 8 +/- 4, an awake SaO2 of 93.1 +/- 3.6 vs 94.1 +/- 2.6 for the rest of the population (p = 0.01). Finally, the false-positive cases had more airways obstruction (FEV1/VC = 72 +/- 13 vs 77 +/- 15%; p = 0.026). Using a delta value of 0.8 leads to a sensitivity of 90% with 19 false-negative cases but with a higher specificity of 75%. A nocturnal oximetry test with a delta index below 0.6 is helpful in ruling out the diagnosis of SAS in patients being screened for this condition, as this yielded only three negative test results in 301 screening procedures.