ArticlePDF Available

Validation of overnight oximetry to diagnose patients with moderate to severe obstructive sleep apnea

Authors:

Abstract and Figures

Polysomnography (PSG) is treated as the gold standard for diagnosing obstructive sleep apnea (OSA). However, it is labor-intensive, time-consuming, and expensive. This study evaluates validity of overnight pulse oximetry as a diagnostic tool for moderate to severe OSA patients. A total of 699 patients with possible OSA were recruited for overnight oximetry and PSG examination at the Sleep Center of a University Hospital from Jan. 2004 to Dec. 2005. By excluding 23 patients with poor oximetry recording, poor EEG signals, or respiratory artifacts resulting in a total recording time less than 3 hours; 12 patients with total sleeping time (TST) less than 1 hour, possibly because of insomnia; and 48 patients whose ages less than 20 or more than 85 years old, data of 616 patients were used for further study. By further considering 76 patients with TST < 4 h, a group of 540 patients with TST ≥ 4 h was used to study the effect of insufficient sleeping time. Alice 4 PSG recorder (Respironics Inc., USA) was used to monitor patients with suspected OSA and to record their PSG data. After statistical analysis and feature selection, models built based on support vector machine (SVM) were then used to diagnose moderate and moderate to severe OSA patients with a threshold of AHI = 30 and AHI = 15, respectively. The SVM models designed based on the oxyhemoglobin desaturation index (ODI) derived from oximetry measurements provided an accuracy of 90.42-90.55%, a sensitivity of 89.36-89.87%, a specificity of 91.08-93.05%, and an area under ROC curve (AUC) of 0.953-0.957 for the diagnosis of severe OSA patients; as well as achieved an accuracy of 87.33-87.77%, a sensitivity of 87.71-88.53%, a specificity of 86.38-86.56%, and an AUC of 0.921-0.924 for the diagnosis of moderate to severe OSA patients. The predictive outcome of ODI to diagnose severe OSA patients is better than to diagnose moderate to severe OSA patients. Overnight pulse oximetry provides satisfactory diagnostic performance in detecting severe OSA patients. Home-styled oximetry may be a tool for severe OSA diagnosis.
Content may be subject to copyright.
R E S E A R C H A R T I C L E Open Access
Validation of overnight oximetry to diagnose
patients with moderate to severe obstructive
sleep apnea
Liang-Wen Hang
1,2
, Hsiang-Ling Wang
3
,Jen-HoChen
4
,Jiin-ChyrHsu
5
, Hsuan-Hung Lin
6
, Wei-Sheng Chung
4,7,8*
and Yung-Fu Chen
4,8,9*
Abstract
Background: Polysomnography (PSG) is treated as the gold standard for diagnosing obstructive sleep apnea (OSA).
However, it is labor-intensive, time-consuming, and expensive. This study evaluates validity of overnight pulse oximetry
as a diagnostic tool for moderate to severe OSA patients.
Methods: A total of 699 patients with possible OSA were recruited for overnight oximetry and PSG examination
at the Sleep Center of a University Hospital from Jan. 2004 to Dec. 2005. By excluding 23 patients with poor
oximetry recording, poor EEG signals, or respiratory artifacts resulting in a total recording time less than 3 hours;
12 patients with total sleeping time (TST) less than 1 hour, possibly because of insomnia; and 48 patients whose
ages less than 20 or more than 85 years old, data of 616 patients were used for further study. By further
considering 76 patients with TST < 4 h, a group of 540 patients with TST 4 h was used to study the effect of
insufficient sleeping time. Alice 4 PSG recorder (Respironics Inc., USA) was used to monitor patients with
suspected OSA and to record their PSG data. After statistical analysis and feature selection, models built based on
support vector machine (SVM) were then used to diagnose moderate and moderate to severe OSA patients with
a threshold of AHI = 30 and AHI = 15, respectively.
Results: The SVM models designed based on the oxyhemoglobin desaturation index (ODI) derived from
oximetry measurements provided an accuracy of 90.42-90.55%, a sensitivity of 89.36-89.87%, a specificity of
91.08-93.05%, and an area under ROC curve (AUC) of 0.953-0.957 for the diagnosis of severe OSA patients; as well
as achieved an accuracy of 87.33-87.77%, a sensitivity of 87.71-88.53%, a specificity of 86.38-86.56%, and an AUC
of 0.921-0.924 for the diagnosis of moderate to severe OSA patients. The predictive outcome of ODI to diagnose
severe OSA patients is better than to diagnose moderate to severe OSA patients.
Conclusions: Overnight pulse oximetry provides satisfactory diagnostic performance in detecting severe OSA
patients. Home-styled oximetry may be a tool for severe OSA diagnosis.
Keywords: Obstructive sleep apnea (OSA), Oximetry, Support vector machine (SVM), Polysomnography (PSG)
* Correspondence: chung.w53@msa.hinet.net;yfchen@ctust.edu.tw
7
Department of Internal Medicine, Taichung Hospital, Ministry of Health and
Welfare, Taichung, Taiwan
8
Department of Healthcare Administration, Central Taiwan University of
Science and Technology, Taichung, Taiwan
Full list of author information is available at the end of the article
© 2015 Hang et al.; licensee BioMed Central. This is an Open Access article distributed under the terms of the Creative
Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and
reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain
Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article,
unless otherwise stated.
Hang et al. BMC Pulmonary Medicine (2015) 15:24
DOI 10.1186/s12890-015-0017-z
Background
Breathing-related sleep disorder is a spectrum of dis-
eases including snoring, upper airway resistance syn-
drome and sleep apnea [1]. Several studies have
explored that sleep apnea is associated with obesity, day-
time fatigue, high blood pressure, and increased risk of
cardiovascular morbidity [2,3]. Obstructive sleep apnea
(OSA) is prevalent in 4% of men and 2% of women [4];
among them, up to 93% of women and 82% of men have
not been diagnosed [5], resulting in an increased risk of
27 folds in causing motor vehicle crashes [6] and caus-
ing several chronic diseases, such as metabolic syndrome
[7]; neurocognitive deficits [8], vigilance alteration and
attentional decline [9]; and erectile dysfunction [10].
Although polysomnography (PSG) is regarded as the
gold standard, it is time-consuming, labor-intensive, and
expensive [11]. According to a recent report, waiting
time for accessing to diagnose and treat patients with
suspected OSA is lengthy even in the developed counties
around Europe, Australia, the United States, and Canada
[12]; for example, it was estimated that the waiting time
for non-urgent referrals for a sleep study ranges from 0
to 48 months in UK, 2 weeks to 2 months in Belgium, 4
to 68 weeks in Australia, a few weeks to more than a
year in the US, and 8 to 30 months in Canada [12]. The
waiting time ranges from 3 to 7 days in our hospital.
Hence, other devices which are cheap, safe, and accur-
ate; readily and easily accessed; and have no risk or side
effects to the patients are needed for decreasing waiting
time and cost for OSA diagnosis [11]. In addition to
labor-intensive, time-consuming, and high examination
cost, PSG also has other limitations, such as technical
expertise is required and timely access is restricted [12].
Thus, home pulse oximetry has been proposed as a valu-
able screening tool, although its effectiveness in screen-
ing patients with OSA has been debated for several years
[13]. A number of studies have assessed its usefulness,
but sometimes with conflicting results [14-16]. For ex-
ample, home overnight oximetry was found to be not
very correlated with PSG for testing children [17] and to
be considerable night-to-night variability for aged pa-
tients with chronic obstructive pulmonary disease
(COPD) [18]. However, Brouillette and colleagues re-
ported that oximetry could be used to diagnose OSA for
children with a positive predictive value of 97% [19].
It was reported that total sleeping time (TST) and
sleep efficiency were significantly smaller for the first
night in-home PSG compared to the 2nd and 3rd nights
[20]. Night-to-night variability of AHI greater than 5
among 21% of the study patients had also been observed
[21]. Furthermore, Newell et al [22] showed that TST,
sleep efficiency, and AHI were significantly different be-
tween 2 consecutive nights of PSG recording for 4 indi-
vidual groups of subjects, including sleep-related
breathing disorders, insomnia, movement and behavioral
disorders, and healthy control, indicating that AHI
might be affected by the TST.
Patients who do not have enough sleep time during
the examination might have great influence on the diag-
nosis of OSA severity. Although Lam and colleagues ex-
cluded subjects with TST less than 4 hours [7], the
reason why these data were removed was not reported.
In this study, we compared two groups of subjects, one
with TST less than 4 hours and another with TST more
than 4 hours.
The support vector machine applied to oximetry for
the diagnosis of OSA has been little explored. Recently,
SVM was used to design classifiers for the diagnosis of
OSA patients based on the data acquired from only 149
and 100 recruited subjects, respectively, by Al-Angari
et al. [23] and Marcos et al. [24]. Marcos et al. [24] de-
signed an SVM classifier using Gaussian kernel to clas-
sify spectral features of SaO
2
signals. The subjects
divided into 60 normal and 89 OSA patients without
considering the severity were recruited in their study.
Al-Angari et al. [23] adopted linear kernel and second-
order polynomial kernel to design the SVM classifiers
for discriminating 1-mimute epochs of OSA segments
from the normal segments, and to discriminate patients
from the normal subjects. The subjects were divided into
50 normal and 50 OSA patients. Only accuracy, sensitiv-
ity, and specificity were evaluated in both studies, while
areas under ROC curve (AUC) were not calculated. In
the present study, the PSG data of 616 patients with
possible OSA were used to design SVM classifiers for
the diagnoses of severe and moderate to severe OSA
patients.
The objectives of this study are to (1) validate noctur-
nal oximetry for the screening of high-risk OSA patients
using oxyhemoglobin desaturation index (ODI) to detect
severe and moderate-to-severe patients; (2) consider the
effect of patients with TST less than 4 hours compared
with those having longer TST during PSG examination;
and (3) determine the best kernel and features used for
constructing the SVM classifiers to discriminate the
OSA patients from the normal subjects.
Methods
Study subjects
A total of 699 patients with possible OSA were recruited
and tested using PSG device for overnight recording at
the Sleep Center of China Medical University Hospital
after written informed consents from Jan. 2004 to Dec.
2005. By excluding 23 patients with total recording time
less than 3 hours, 12 with poor oximetry recordings,
poor EEG signals, or respiratory artifacts, and 48 sub-
jects whose ages less than 20 or more than 85 years old
(7), data of 616 patients were used for analysis. By
Hang et al. BMC Pulmonary Medicine (2015) 15:24 Page 2 of 13
further considering 76 patients with TST less than
4 hours [7], a group of 540 patients whose TST more
than 4 hours was used to study the effect of insufficient
sleeping time. The study has been approved by Institu-
tional Review Board of China Medical University (Study
No. DMR 96-IRB-17).
PSG examination
Alice 4 PSG recorder (Respironics Inc., USA) was used
to monitor and record patients with suspected OSA,
during which a number of physiologic variables were
measured and recorded during sleep. Physiologic sensors
were used to record (1) EEG for detecting brain elec-
trical activity and sleep staging on the basis of 30-sec
epochs, (2) EOG and submental EMG for detecting eye
and jaw muscle movement, (3) tibia EMG for monitor-
ing leg muscle movement, (4) airflow (by using nasal
cannula to measure oronasal pressur) for detecting
breath interruption (1287 nasal flow, PTAF 2; Pro-Tech;
Pittsburgh, PA), (5) inductance plethysmorgraphy for es-
timating respiratory effort (chest and abdominal excur-
sion), (6) ECG for measuring heart rate, and (7) arterial
oxygen saturation by using a finger probe oximeter (935
Oximeter Sensor; Respironics; Murrysville, PA).
PSG recordings were performed overnight (from
9:00 PM to 06:00 AM in general) in the sleep center for
a minimum of 3 hours of light-on to light-off recording
time when patients received OSA diagnoses. During the
examination, the attending technicians on duty closely
inspected the acquired signals of individual channels dis-
played on the monitor to check their signal quality. If
the signal quality was poor or contained obvious arti-
facts, the technicians would check the instrument setup,
cable connection, and electrode-skin contact on-site to
ensure proper recording. The quality of PSG raw data
was verified by an experienced physician of sleep medi-
cine and scored by a technician certificated by the
Taiwan Society of Sleep Medicine. During off-line ana-
lysis, the periods with artifacts or poor signal quality not
meeting the diagnostic requirements, such as poor sig-
nals or artifacts in acquired oximetry, EEG, or respira-
tory data that did not allow the reading of the SpO2,
sleep stages, or the respiratory events, were deleted from
the recorded signals. For patients whose net total re-
cording time less than 3 hours or TST less than 1 hour
were arranged to receive another free examinations.
Sleeping efficiency was calculated as TST divided by
total recording time.
The oximeter was used to measure the SpO
2
with a
sampling rate of 1 Hz. Only the SpO
2
signal acquired
from pulse oximeter was used for ODI calculation with-
out the knowledge of other PSG information. For pulse
rates less than 112 beat/min, the SpO
2
calculation is
based on a four-beat exponential averaging. The
averaging time is doubled to 8 beats for pulse rate
greater than 112 beat/min, and is redoubled again for
pulse rate more than 225 beat/min. Artifacts with
changes of oxygen saturation between consecutive sam-
pling intervals greater than 4%/s, and any oxygen satur-
ation less than 50% were detected and eliminated by an
automated algorithm [25].
An apnea was defined as a minimum of 10 seconds of
airflow cessation, and a hypopnea was determined by a
30% reduction in airflow preceding a period of normal
breathing for a minimum of 10 seconds and oxyhemo-
globin desaturation (decrease in SpO
2
4%) [26].
Apnea-hypopnea index (AHI) was calculated as the total
number of apneas (central, mixed, or obstructive) and
hypopneas per hour of sleep. An arousal event was
defined as an episode lasting 3 seconds or more with a
return of αactivity associated with a detectable increase
in EMG activity [27]. Arousal index (AI) was the total
number of arousals per hour of sleep. OSA patients were
diagnosed as mild, moderate, and severe with an AHI
greater than 5/h, 15/h, and 30/h, respectively.
The questionnaires including demographic informa-
tion (age, gender, education, profession type, as well as
smoking and drinking status), Epworth scaling score
(ESS), as well as symptoms (snoring, drooling and teeth-
grinding during sleep; feeling headache, tired, and dry
mouth after wake-up; daytime drowsiness, memory
problem, and traffic accidence caused by drowsiness
when driving) and comorbidities (hypertension, diabetes,
high blood lipid, high uric acid, stroke, heart attack, angina
pectoris, asthma, thyroid disorder, and allergic rhinitis) re-
lated to OSA diagnosis were filled by the patients before
PSG recording. Anthropometric parameters (weight,
height, as well as waist, neck and hip circumferences) were
measured and BMI calculated by the technicians.
Oxyhemoglobin desaturation index derived from pulse
oximetry measurement
Oxyhemoglobin desaturation index (ODI), represented
as ODInb, includes three components, i.e., threshold (n),
baseline (b), and lasting time (fixed to t=3 sec in this
study) parameters. For threshold n, the amount of oxy-
hemoglobin desaturation n% decrease from the baseline
was calculated. In this study, two different baselines, in-
cluding the mean of all-night oxygen (A) [28] and the
mean of the top 20% of oxyhemoglobin saturation values
over the 1 min preceding the scanned oxyhemoglobin
value (T) [16] were adopted. The lasting time parameter
t= 3 was defined as that an oxyhemoglobin desaturation
event with a minimum of 3 sec were continued. Finally,
the ODI was calculated by dividing the total oxyhemo-
globin desaturation counts with the total recording time
(hours). For example, ODI4A calculated the amount of
oxyhemoglobin desaturation events with a 4% decrease
Hang et al. BMC Pulmonary Medicine (2015) 15:24 Page 3 of 13
from the mean of all-night oxygen lasting for more than
3 seconds per hour. Because the oximetry is one of the
PSG channels, we aim to investigate the accuracy, sensi-
tivity, specificity, and area under ROC curve (AUC) of
overnight oximetry for the diagnosis of moderate and
moderate to severe OSA patients by ODI parameters, in-
cluding ODI2 (ODI2T), ODI3 (ODI3T), ODI4T, and
ODI4A, derived from oximetry measurements.
Consideration of TST during PSG examination
Subjects who do not have enough sleep time during the
examination might have great influence on the diagnosis
of OSA severity. Although Lam and colleagues excluded
subjects with TST less than 4 hours [7], the reason why
these data were removed was not reported. In this study,
we compared two groups of subjects, one with TST less
than 4 hours and another one with TST more than
4 hours.
Analytical techniques
Demographic, anthropometric, questionnaire, and PSG
data of patients were analyzed using descriptive statis-
tical analysis for calculating means and standard devia-
tions of individual variables. Inferential statistical
analyses including Pearson chi-square test and analysis
of variance (ANOVA) were applied to detect significant
variables for discriminating among normal subjects and
mild, moderate, and severe OSA patients. Moreover, linear
regression analysis and Bland-Altman plot were per-
formed to compare correlations and agreements between
ODI variables (ODI2, ODI3, ODI4T, and ODI4A) and
AHI. Analysis of ROC curves were performed to set the
cutoff values of different ODI parameters that best dis-
criminate moderate and moderate/severe OSA patients
from normal and mild subjects. Furthermore, we tested
and compared various combinations of neck circumfer-
ence (NC), body mass index (BMI), ESS, and ODI vari-
ables in the diagnosis of severe and moderate to severe
patients based on 2 thresholds, AHI = 30 and AHI = 15,
respectively. Finally, support vector machine (SVM), an
artificial technique, was used to design computer-assisted
diagnostic systems based on parameters derived from the
oximetry measurements only by excluding signals ac-
quired from other PSG channels. Comparisons of the
diagnostic performance of traditional oximetry parameters
with those from the SVM models were also conducted.
Support vector machine
The SVM technique is a useful technique for data classi-
fication and regression, which has become an important
tool for machine learning and data mining. In general,
SVM has better performance when competed with exist-
ing methods, such as neural networks and decision trees
[29-31]. Recently, application of SVM in medicine has
grown rapidly. For examples, it has been applied in pre-
diction of disulfide bonding patterns in proteins [32],
discrimination of malignant and benign cervical lymph
nodes [33], disease diagnosis using tongue images [34],
diagnoses of cardiovascular disease [35] and breast cancer
[36], and predictions of successful ventilation weaning
[37]. A special property of SVM is that it can simultan-
eously minimize the empirical classification error and
maximize the geometric margin. Its goal is to separate
multiple clusters with a set of unique hyperplanes having
greatest margin to the edge of each cluster to reduce
misclassification induced by noise, where each hyper-
plane separating two clusters is not unique for ordinary
linear classifiers. Recently, SVM was also used to design
classifiers for the diagnosis of OSA patients [23,24].
Results
Demographic and clinical characteristics of participants
The tested subjects were divided into 4 groups based on
the AHI for statistical analysis. Table 1 shows the demo-
graphic and clinical characteristics of participants and
compares the statistic results among 4 patient groups.
Regarding the influence of gender in OSA severity, the
proportion of male patients increases significantly fol-
lowing the OSA severity (Chi-square test, p< 0.001).
Demographic and anthropometric variables (age, BMI,
and NC), ESS obtained from the questionnaire, and vari-
ables derived from oximetry (ODI and heart rate), reach-
ing a level of significance (ANOVA, p< 0.01) among two
or three stages, were selected as potential variables for
designing the SVM classifiers to detect severe (30AHI)
and moderate to severe (15AHI) patients. It was found
that only ODI could be used to differentiate all 4 differ-
ent groups. Variables, including TST, sleep latency,
arousal count, and arousal index, derived from EEG
and/or EMG, were also calculated and compared among
4 groups of participants. It can be observed that the TST
is lower while the arousal count and arousal index are
higher for more severe patients.
Discrimination of normal, mild, moderate, and severe
participants
Table 2 shows the confusion matrices of multiclass clas-
sifiers in the discrimination of 4 groups of participants
using ODI parameters derived from the oximetry mea-
surements for two datasets including 616 (dataset 1) and
540 (dataset 2) subject data, respectively. ODI parame-
ters can provide a total predictive accuracy of 71.27%
and 73.52% for dataset 1 and dataset 2, respectively. The
sensitivities of SVM models in diagnosing severe pa-
tients are 88.07% and 87.76% in dataset 1 and dataset 2,
respectively, while the sensitivities are less than 75% for
other 3 categories. The predictive accuracy is obtained
by tenfold cross validation.
Hang et al. BMC Pulmonary Medicine (2015) 15:24 Page 4 of 13
Diagnosis of severe and moderate/severe OSA patients
Table 3 displays various combinations of NC, BMI, ESS,
and ODI for designing predictive models in the diagnosis
of severe OSA patients, and the prediction rates and AUC
achieve 89.81-90.58% and 0.946-0.958, respectively. Diag-
nostic models designed using ODI parameters derived
from pulse oximetry alone achieved prediction perform-
ance similar to other multiclass predictors designed using
more variables. Models designed with ODI parameters
provided a sensitivity of 87.36-89.87%, a specificity of
91.08-93.05%, an accuracy of 90.42-90.55%, and an AUC
of 0.953-0.957 for the diagnosis of severe OSA patients in
both datasets.
Table 4 compares the diagnostic performance of various
diagnostic models designed with different combination of
variables for the diagnosis of moderate to severe patients.
Table 2 Confusion matrices for the classification of 4 groups with ODI used as predictor for two datasets containing
616 and 540 samples, respectively
N Normal Mild Moderate Severe Sensitivity (%)
Dataset 1
Normal 72 53 10 1 0 73.61
Mild 127 25 80 20 2 62.99
Moderate 132 3 48 55 26 41.67
Severe 285 1 11 22 251 88.07
Total Validation 616 Overall accuracy: 71.27%
Dataset 2
Normal 68 51 17 0 0 75.00
Mild 121 21 85 14 1 70.25
Moderate 114 3 39 53 19 46.49
Severe 237 0 10 19 208 87.76
Total Validation 540 Overall accuracy: 73.52%
Table 1 Statistic tests of demographic, questionnaire, and PSG data obtained from patients across four stages of
severity based on AHI value
Severity
a
All (1) Normal (2) Mild (3) Moderate (4) Severe Statistic test
b
Patient No. (%) 616 (100%) 72 (11.7%) 127 (20.6%) 132 (21.4%) 285 (46.3%)
Gender (%)
Female 141 (22.9%) 32 (44.4%) 49 (38.6%) 31 (23.5%) 29 (10.2%) χ
2
= 62.801, p< 0.001
Male 475 (77.1%) 40 (55.6%) 78 (61.4%) 101 (76.5%) 256 (89.8%)
Age
***
45.2 ± 12.7 36.2 ± 10.2 42.1 ± 11.5 45.9 ± 12.8 48.4 ± 12.5 (1) < (2),(3),(4); (2) < (4)
BMI
***
26.74 ± 4.24 23.88 ± 3.15 25.60 ± 3.75 26.18 ± 3.68 28.22 ± 4.33 (1),(2),(3) < (4)
NC
***
39.08 ± 3.65 36.24 ± 3.25 37.53 ± 3.52 38.69 ± 3.19 40.79 ± 3.10 (1) < (3); (1),(2),(3) < (4)
ESS
**
9.21 ± 5.24 7.67 ± 5.16 8.54 ± 4.77 9.19 ± 5.25 9.93 ± 5.34 (1) < (4)
ODI parameter
ODI2
***
37.58 ± 23.50 10.25 ± 7.18 19.81 ± 10.57 29.60 ± 12.31 56.11 ± 18.94 (1) < (2) < (3) < (4)
ODI3
***
26.16 ± 23.93 3.10 ± 2.78 8.18 ± 6.02 15.54 ± 8.87 44.92 ± 22.37 (1) < (2) < (3) < (4)
ODI4T
***
20.44 ± 23.06 1.28 ± 1.43 4.18 ± 3.80 9.52 ± 7.01 37.58 ± 23.66 (1) < (2) < (3) < (4)
ODI4A
***
24.65 ± 26.00 1.14 ± 1.41 5.20 ± 3.94 12.01 ± 7.62 45.10 ± 24.97 (1) < (2) < (3) < (4)
Heart Rate 73.32 ± 11.02 70. 31 ± 8.45 71.05 ± 10.61 72.64 ± 9.50 75.41 ± 11.97 (1),(2) < (4)
TST (min)
***
309.0 ± 63.4 317.8 ± 56.3 331.3 ± 48.9 307.6 ± 58.5 297.5 ± 69.8 (3),(4) < (2)
Latency (min)
***
19.6 ± 21. 1 27. 6 ± 33.0 16.2 ± 13.6 22.4 ± 23.2 17.9 ± 18.2 (2),(4) < (1)
Arousal count
***
153.9 ± 99.1 95.3 ± 55.6 103.0 ± 52.3 121.1 ± 52.3 206.5 ± 112.4 (1),(2),(3) < (4)
Arousal Index
***
33.6 ± 20.7 19.3 ± 10.8 20.6 ± 10.1 26.6 ± 11.2 46.3 ± 21.9 (1),(2),(3) < (4); (1) < (3)
a
Normal: AHI < 5; Mild: 5AHI < 15; Moderate: 15AHI < 30; Severe: AHI30.
b
ANOVA test with
*
p< 0.05,
**
p< 0.01, and
***
p< 0.001; Pearson Chi-squa re test with
p< 0.001.
Hang et al. BMC Pulmonary Medicine (2015) 15:24 Page 5 of 13
The sensitivity, specificity, accuracy, and AUC achieve
87.71-89.97%, 83.08-86.56%, 86.85-88.14%, and 0.921-0.941,
respectively. Again, ODI alone provides a similar predictive
accuracy of 87.33-87.77%, a sensitivity of 87.71-88.53%, a
specificity of 86.38-86.56%, and an AUC of 0.921-0.924, re-
spectively, for the diagnosis of moderate to severe OSA
patients.
Figure 1 illustrates and compares the ROC curves of
models designed using traditional ODI variables (ODI2,
ODI3, ODI4T, and ODI4A) and their combinations
(ODI2 and ODI4A). In addition to higher accuracy
(Table 3), the AUCs of the SVM models designed using
2 ODI variables are greater than other single-variable
models, as shown in Figure 1(a) and 1(b), indicating its
capability in diagnosing severe OSA patients outper-
forms other traditional models. Regarding the diagnosis
of moderate to severe patients, as shown in Table 4, the
multiple-variable SVM models and traditional single-
variable models exhibit similar diagnostic performance.
Cutoff values for diagnosis of OSA patients using ODI
variables
The cutoff values of single-variable models for the diag-
nosis of severe and moderate to severe OSA patients are
shown in Tables 3 and 4, respectively. The optimal cutoff
values for severe patient diagnosis are 27.2, 18.4, 11.2,
and 13.7, respectively, in dataset 1, as well as 27.2, 18.4,
10.4, and 13.7, respectively, in dataset 2 for ODI2, ODI3,
ODI4T, and ODI4A. The cutoff values were decreased
to 21.2, 9.5, 7.3, and 8.5 for dataset 1 as well as 21.2, 9.5,
7.3, and 7.3 for dataset 2 when making diagnosis of
moderate to severe patients using ODI2, ODI3, ODI4T,
and ODI4A, respectively.
Effect of less TST
Table 5 shows that the number and percentage of sub-
jects with different OSA severity, exhibiting significant
difference between patients with TST less than 4 hours
and those with TST more than 4 hours (p< 0.01). As
Table 3 Diagnosis of severe patients with AHI =30 as the threshold using different combination of salient features
Data Multiple variables Single variable
Predictive index (%) NC, BMI, ESS, ODI ODI, ESS ODI, BMI ODI ODI2 ODI3 ODI4T ODI4A
Dataset 1 (N = 616) Accuracy 90.42 90.58 90.58 90.42 87.2 89.5 87.8 89.8
Sensitivity 88.07 87.01 86.31 87.36 91.6 86.0 85.7 90.6
Specificity 92.44 93.65 94.25 93.05 83.3 92.4 89.7 89.1
AUC 0.958 0.956 0.954 0.957 0.938 0.942 0.912 0.913
Cutoff - - - - 27.2 18.4 11.2 13.7
Dataset 2 (N = 540) Accuracy 90.18 90.37 89.81 90.55 87.6 90.0 88.5 89.3
Sensitivity 86.07 88.18 88.60 89.87 91.6 86.6 88.7 89.5
Specificity 93.39 92.07 90.75 91.08 84.4 92.7 88.4 89.1
AUC 0.952 0.946 0.950 0.953 0.945 0.946 0.913 0.913
Cutoff - - - - 27.2 18.4 10.4 13.7
Note: ODI is the combination of ODI2 and ODI4A.
Table 4 Diagnosis of moderate to severe patients with AHI = 15 as the threshold using different combination of salient
features
Data Multiple variable Single variable
Predictive index (%) NC, BMI, ESS, ODI ODI, ESS ODI, BMI ODI ODI2 ODI3 ODI4T ODI4A
Dataset 1 (N = 616) Accuracy 86.85 87.82 87.01 87.33 87.2 89.5 87.8 89.8
Sensitivity 88.67 89.87 87.95 87.71 91.6 86.1 85.7 90.6
Specificity 83.08 83.58 85.07 86.56 83.3 92.4 89.7 89.1
AUC 0.938 0.935 0.927 0.921 0.918 0.920 0.870 0.869
Cutoff - - - - 21.2 9.5 7.3 8.5
Dataset 2 (N = 540) Accuracy 87.96 88.14 87.59 87.77 87.6 90.0 88.5 89.3
Sensitivity 89.39 89.97 88.82 88.53 91.6 86.6 88.7 89.5
Specificity 85.34 84.81 85.34 86.38 84.4 92.7 88.4 89.1
AUC 0.941 0.939 0.940 0.924 0.925 0.927 0.872 0.875
Cutoff - - - - 21.2 9.5 7.3 7.3
Note: ODI is the combination of ODI2 and ODI4A.
Hang et al. BMC Pulmonary Medicine (2015) 15:24 Page 6 of 13
indicated in the table, the percentage of severe patients
in the group of TST < 4 h group (63.2%) is higher than
the TST 4 h group (43.9%), while it is opposite by con-
sidering the cumulated normal and mild subjects (13.2%
vs 35.0%). On the other hand, the percentages of moder-
ate patients are very close between two groups (23.9% vs
21.1%). No significant difference (p> 0.05) was observed
between two groups regarding anthropometric variables
(BMI and NC), questionnaire (ESS), and HR, while sig-
nificant difference was found for age (p< 0.001), ODI4A
(p<0.05), and AHI (p< 0.01). Interestingly, except
ODI4A, other ODI parameters, i.e., ODI2, ODI3, and
ODI4T, exhibited no significant difference (p>0.05) be-
tween two different TST groups. It mimics ODI4A is more
sensitive than ODI2, ODI3, and ODI4T in discriminating
patients with less TST from those with more TST. Fur-
thermore, as shown in Table 5, patients with less TST
demonstrate significantly greater latency (p<0.001)
and arousal index (p<0.05);however,thearousal
counts is significantly smaller (p< 0.001). We suspected
that data recorded from subjects who do not have
enough sleeping time might not be reliable for analysis
and should be excluded. An additional PSG examination
should be conducted to draw more solid diagnostic
conclusion.
Discussion and conclusion
Prediction of OSA using questionnaires, demographics,
clinical features, and physiological examination has been
extensively studied in the last decade [11,38]. Goncalves
and colleagues used Epworth sleeping scale (ESS), the
sleeping disorders questionnaire, the Beck depression in-
ventory, the medical outcome study 36-item short form
health survey (SF-36), and a questionnaire on driving
difficulties and accidents to evaluate subjects who were
suspected to have breathing-related sleep disorders.
Among them, ESS was found to be correlated to arousal
index and apnea-hyponea index (AHI) [38], while con-
tradicted by other investigators [11,39]. Khoo et al [40]
used questionnaires, containing questions regarding
snoring, choking, suffocating, and abrupt awaking during
sleep, to study Asian populations, including Chinese,
Malaysian, and Indian. It was found that the risk factors
are similar to white populations in strong association of
snoring and sleep apnea with male gender, older age,
obesity, family history, and smoking [40]. In addition,
demographic, clinical, and biochemical factors including
age, sex, observed sleep apnea, fasting insulin, glycosylated
hemoglobin A1
C
, and central obesity and lager NC were
found to significantly increase the risk of higher AHI for
the severely obese patients [39]. Strong correlation be-
tween patient self-perception and clinical examination, in-
cluding Friedman tongue position grade and Friedman
clinical staging, of OSA severity and AHI was also found
[11]. In this study, ESS questionnaire was considered to be
included for designing SVM classifiers for OSA diagnosis.
As indicated in Table 1, it is only effective in discriminat-
ing the normal from the severe OSA patients. In contrast,
all ODI parameters are able to discriminate the normal
and different stages of OSA patients. As presented in Ta-
bles 3 and 4, diagnostic performance is not improved
when ODI parameters are combined with ESS, nor is
combined with BMI and/or NC. ODI parameters provided
by overnight oximetry measurements may become good
Figure 1 ROC curves of ODI parameters for the diagnosis of
severe and moderate/severe OSA patients with thresholds (a)
AHI=30 and (b) AHI=15, respectively.
Hang et al. BMC Pulmonary Medicine (2015) 15:24 Page 7 of 13
predictors in the diagnosis of OSA, while demographic
and questionnaire variables are not very helpful to elevate
the prediction rate [41].
Selection of ODI parameters for designing SVM models
The selection of variables for designing predictive
models for OSA diagnosis is based on ANOVA analysis,
linear regression analysis, and Bland-Altman plot. Al-
though all variables shown in Table 1 reached a level of
significance in discriminating 4 groups of participants
when tested with ANOVA, only NC, BMI, and ODI pa-
rameters were considered as potential predictors after
post-hoc analysis in discriminating severe patients from
the other groups. In contrast, all the ODI parameters are
able to discriminate any 2 groups of participants. The
variables derived from PSG channels other than oxim-
etry, including TST, sleep latency, arousal count, and
arousal index, were not included, because they cannot
be derived from the home oximetry.
Several ODI parameters, including ODI2, ODI3, ODI4T,
and ODI4A, which were highly correlated with AHI were
considered to be included for designing the SVM models.
In addition to correlation, their agreements with AHI were
also evaluated by Bland-Altman plots. After linear regres-
sion analysis, the obtained linear predictor functions were
used to perform Bland-Altman plots to check their agree-
ments with the gold standard AHI. As illustrated in
Figure 2, Bland-Altman plots were compared among single
and combined ODI parameters after linear regressions.
The correlation coefficients (R
2
) of AHI versus ODI2,
ODI3, ODI4T, and ODI4A were 0.770, 0.835, 0.813, and
0.878, respectively. The agreements of AHI with individual
ODI parameters were evaluated by observing their means
and differences. As exhibited in Figure 2 (a)-(d), the 95%
limits of agreements (±1.96 SD) are ±26.4, ±22.4, ±23.8,
and ±19.2, respectively, for ODI2, ODI3, ODI4T, and
ODI4A with a mean of 0. ODI4A presented best agree-
ment with AHI and was defined as the potential variable
for combining with other variables to achieve better diag-
nostic performance. Figure 2 (e) and (f ) demonstrate
the Bland-Altman plots for ODI4A combined with
ODI2 and ODI3, respectively. It can be found that the
former combination (R
2
) = 0.881, ±1.96 SD = ±19.0)
exhibits slightly better correlation and agreement with
AHI than the latter (R
2
= 0.880, ±1.96 SD = ±19.1) and
ODI4A alone (R
2
= 0.878, ±1.96 SD = ±19.2). The cor-
relations and agreements were not improved by in-
cluding more ODI parameters; for example, as shown
in Figure 3, the correlation and agreements were
slightly degraded for the combined ODI4A, ODI2, and
Table 5 Comparison of statistic results of demographic, anthropometric, questionnaire, and PSG variables between
patients whose TST < 4 h and those with TST 4h
Severity
a
All TST < 4 h TST 4 h Statistic test (p)
b
AHI
**
34.07 ± 28.11 43.52 ± 26.81 32.74 ± 28.06 0.002
Severity
616 (100%) 76 (12.34%) 540 (87.66%)
Normal 72 (11.7%) 4 (5.3%) 68 (12.6%) χ
2
= 15.4
Mild 127 (20.6%) 6 (7.9%) 121 (22.4%) p< 0.01
Moderate 132 (21.4%) 18 (23.9%) 114 (21.1%)
Severe 285 (46.3%) 48 (63.2%) 237 (43.9%)
Age
**
45.2 ± 12.7 52.20 ± 14.67 44.17 ± 12.16 0.001
BMI 26.74 ± 4.24 27.14 ± 4.37 26.68 ± 4.22 0.370
NC 39.08 ± 3.65 39.65 ± 3.56 39.06 ± 3.42 0.161
ESS 9.21 ± 5.24 9.43 ± 5.48 9.19 ± 5.22 0.599
ODI parameter
ODI2 37.58 ± 23.50 41.48 ± 19.42 37.03 ± 23.98 0.073
ODI3 26.16 ± 23.93 27.53 ± 19.18 25.96 ± 24.53 0.521
ODI4T 20.44 ± 23.06 19.84 ± 17.99 20.53 ± 23.70 0.767
ODI4A
*
24.65 ± 26.00 31.16 ± 24.23 23.73 ± 26.13 0.015
Heart Rate 73.32 ± 11.02 73.85 ± 12.49 73.24 ± 10.79 0.703
TST (min)
**
309.0 ± 63.4 180.2 ± 53.3 327.1 ± 39.1 0.002
Latency (min)
***
19.6 ± 21. 1 41.5 ± 40.2 16.6 ± 14.4 <0.001
Arousal count
***
153.9 ± 99.09 103.2 ± 65.3 161.0 ± 101.0 <0.001
Arousal Index
*
33.6 ± 20.7 39.3 ± 22.4 32.8 ± 20.3 0.011
a
Normal: AHI < 5; Mild: 5AHI < 15; Moderate: 15AHI < 30; Severe: AHI30.
b
ANOVA test with
*
p< 0.05,
**
p< 0.01, and
***
p< 0.001; Pearson Chi-squa re test with
p< 0.001.
Hang et al. BMC Pulmonary Medicine (2015) 15:24 Page 8 of 13
-60
-40
-20
0
20
40
60
80
yODI2=1.05xODI2-5.39, R
2
=0.770
-10 10 30 50 70 90 110 130 150 170 190
Mean of AHI and yODI2
AHI - yODI2
AHI - yODI3
Mean
-0.0
-1.96 SD
-26.4
+1.96 SD
26.4
-60
-40
-20
0
20
40
60
80
yODI3=1.074xODI3+5.987, R
2
=0.835
0 20 40 60 80 100 120 140 160 180
Mean of AHI and yODI3
Mean
-0.0
-1.96 SD
-22.4
+1.96 SD
22.4
-60
-40
-20
0
20
40
60
80
yODI4T=1.099xODI4T+11.599, R
2
=0.813
0 20 40 60 80 100 120 140 160 180
Mean of AHI and yODI4T
AHI - yODI4T
AHI - yODI4A
Mean
0.0
-1.96 SD
-23.8
+1.96 SD
23.8
-60
-40
-20
0
20
40
60
80
yODI4A=1.014xODI4A+9.091, R
2
=0.878
0 20 40 60 80 100 120 140 160 180
Mean of AHI and yODI4A
Mean
-0.0
-1.96 SD
-19.2
+1.96 SD
19.2
-60
-40
-20
0
20
40
60
80
yODI4A_2=0.881xODI4A+0.161xODI2+6.312, R
2
=0.881
0 20 40 60 80 100 120 140 160 180
Mean of AHI and
y
ODI4A_2
AHI - yODI4A_2
AHI - yODI4A_3
Mean
-0.0
-1.96 SD
-19.0
+1.96 SD
19.0
-60
-40
-20
0
20
40
60
80
yODI4A_3=0.86xODI4A+0.173xODI3+8.349, R
2
=0.880
0 20 40 60 80 100 120 140 160 180
Mean of AHI and
y
ODI4A_3
Mean
-0.0
-1.96 SD
-19.1
+1.96 SD
19.1
(a) (b)
(c) (d)
(e) (f)
Figure 2 Bland-Altman plots of AHI vs ODI parameters after linear regression analysis in dataset 1: (a) yODI2=1.05×ODI2-5.39,
R
2
=0.770; (b) yODI3=1.074×ODI3+5.987, R
2
=0.835; (c) yODI4T=1.099×ODI4T+11.599, R
2
=0.813; (d) yODI4A=1.014×ODI4A+9.091,
R
2
=0.878; (e) yODI4A_2=0.881×ODI4A+0.161×ODI2+6.312, R
2
=0.881; and (f) yODI4A_3=0.86×ODI4A+0.173×ODI3+8.349, R
2
=0.882.
Hang et al. BMC Pulmonary Medicine (2015) 15:24 Page 9 of 13
ODI3 (R
2
= 0.881 and ±1.96 SD = ±20.0) as well as
for the combined ODI4A, ODI4T, ODI2, and ODI3
(R
2
= 0.880 and ±1.96 SD = ±38.9) with significantly
deviated means of 3.5 and 15.4, respectively. The
analysis of ROC curves further validated that the
combination of ODI4A and ODI2 achieved better
diagnostic performance, especially for the diagnosis
of severe OSA patients, than the predictors based on
single ODI parameters.
Comparison of multi-variable SVM classifiers and single-
variable predictors
As shown in the confusion matrices (Table 2) resulted
from multiclass SVM classifiers for the detection of nor-
mal, mild, moderate, and severe patients, the sensitivity of
ODI variables (ODI 2 and ODI4A) achieves more than
87% in detecting severe OSA patients, while it is only 41-
74% for the other three groups. The misclassification rates
of severe and moderate patients classified as normal are
0.35% (1/285) and 2.27% (3/132), respectively, for dataset
1, as well as 0% and 2.63% (3/114), respectively, for dataset
2. The misclassified rates are only 1.38% and 0%, respect-
ively, for normal subjects to be classified as moderate or
severe when tested with dataset 1 and dataset 2. The
results mimic that a system designed based on ODI
parameters for diagnosing moderate or moderate/severe
patients is expected to have great diagnostic performance.
The sensitivity in detecting severe patients is 87.36%,
which indicates that 12.64% of the severe patients will be
treated as normal, mild, or moderate (Table 3). Further
PSG test will be expected to confirm the diagnoses of
those mild and moderate patients. On the other hand,
the percentage of normal, mild and moderate patients
who were diagnosed as severe is 6.95% (1-specificity).
According to the results of our dataset indicated in
Table 2, none of the normal subject was diagnosed as
severe, mimicking that the subjects being diagnosed as
severe were mild or moderate OSA patients. It is accept-
able for this misclassification (false-positive) since it was
suggested that mild and moderate patients also need
treatments using continuous positive airway pressure
(CPAP) [42]. The oximetry can be effective to be used
for diagnosing severe OSA patients. The predictive
model is suitable to predict severe patients using
cheaper and convenient oximetry, while the non-severe
patients, including normal, mild, and moderate patients,
need to be confirmed by more expensive PSG examina-
tions. Cost-effectiveness analysis needs to be conducted
to evaluate the strategy of home oximetry test followed
by PSG examination for the diagnosis of OSA patients
to reduce healthcare cost and medical facility usage.
Marcos et al. designed an SVM classifier using Gaussian
kernel to classify spectral features of SaO
2
signals [24]. The
149 subjects were divided into 60 normal and 89 OSA
patients. The severity of OSA patients was not considered
in their study. Since Gaussian kernel is deemed as an
example of RBF kernel, the results were compared with
our results obtained from RBF SVM classifier. The
obtained accuracy, sensitivity, and specificity were 88.00%,
84.44%, and 93.33%, respectively, with an AUC of 0.921.
In comparison, as presented in Tables 3, our results
showed better performance with accuracy (90.42%), sensi-
tivity (87.36%), specificity (91.08%), and AUC (0.953) in
diagnosing severe OSA patients; and, as indicated in Table 4,
similar performance with accuracy (87.33%), sensitivity
(87.71%), specificity (86.38%), and AUC (0.921) in diag-
nosing moderate to severe patients.
-60
-40
-20
0
20
40
60
80
0 20 40 60 80 100 120 140 160 180
Mean of AHI and AHIODI4A_3_2
AHI - AHI ODI4A_3_2
Mea n
3.5
-1.96 SD
-16.5
+1.96 SD
23.5
-100
-80
-60
-40
-20
0
20
40
0 20 40 60 80 100 120 140 160 180 200 220 240
Mean of AHI and yODI4A_4T_3_2
AHI - yODI4A_4T_3_2
Mean
-15.4
-1.96 SD
-54.2
+1.96 SD
23.5
ab
yODI4A_3_2=0.888xODI4A-0.150xODI3+0.168xODI2+6.252, R
2
=0.881
yODI4A_4T_3_2=0.914xODI4A-0.376xODI4T+0.477xODI3-0.004xODI2+6.898, R 2=0.882
Figure 3 Bland-Altman plots of AHI vs ODI parameters after linear regression analysis in dataset 1: (a) yODI4A_3_2=0.888×ODI4A-
0.150×ODI3+0.168×ODI2+6.252, R
2
=0.881 and (b) yODI4A_4T_3_2=0.914×OD I4A-0.376×ODI4T+0.477×ODI3-0.004×ODI2+6.898, R
2
=0.882.
Hang et al. BMC Pulmonary Medicine (2015) 15:24 Page 10 of 13
Al-Angari et al. adopted linear kernel and second-
order polynomial kernel to design the SVM classifiers
for discriminating 1-mimute epochs of OSA segments
from the normal segments, and to discriminate patients
from the normal subjects [23]. Only 100 subjects (50
normal subjects and 50 OSA patients) were recruited in
this study, lacking the representativeness of the study
samples. The best accuracies achieved for the SVM clas-
sifiers designed based on the oxygen saturation features
are 80.10% (Sensitivity: 60.9%; Specificity: 94.1%) and
95% (Sensitivity: 100%; Specificity: 90.2%) for epoch clas-
sification and subject classification, respectively. No
AUC value was provided for both classifications.
Schlotthauer et al. [43] adopted a novel method based
on the empirical mode decomposition to detect oxy-
hemoglobin desaturation events from the recorded pulse
oximetry signals. ODI was then calculated from the oxy-
hemoglobin desaturation events to detect moderate
OSA syndrome with a cutoff ODI value of 18.521,
achieving a sensitivity, specificity, and AUC of 0.851,
0.853, and 0.923, respectively. Its diagnostic performance
is slightly worse than the predictor ODI3 and the SVM
model designed using combined ODI parameters, i.e.,
ODI2 and ODI4A, but is similar to the single-variable
predictors, ODI2, ODI4T, and ODI4A (Table 4).
As shown in Table 6, diagnostic performance of SVM
models embedding different kernels used for the diagno-
sis of severe OSA patients is compared. SVM models
embedding RBF kernels present slightly better diagnostic
performance than models designed with linear or poly-
nomial kernels.
Effect of less TST
By considering the TST, as depicted in Table 5, age,
ODI, and AHI exhibit significant difference between pa-
tients with TST less than 4 hours and those with TST
more than 4 hours. It indicates that older subjects and
more severe OSA patients with higher ODI and AHI
tend to have less sleeping time. There are two possible
explanations of this finding: (1) some severe patients
tend to sleep less because of frequent apnea or hypopnea
occurrences and (2) data recorded from subjects who do
not have enough sleeping time might not be reliable for
further analysis. Regarding the first possibility, the fre-
quent occurrences of apnea or hypopnea after having
fallen asleep induces insomnia for severe patients. The
environmental change might also be the reason for caus-
ing insomnia [44]. By observing the latency time in Table 5,
significant difference (p< 0.001) can be found between
two groups. Subjects with less sleeping time demonstrate
greater latency. Although arousal count for TST less than
4 hours group is significantly smaller than those with TST
more than 4 hours group (p< 0.001); however, the arousal
index is significantly greater (p< 0.01). We suggest that
subjects who were diagnosed as normal but didnttake
enough sleeping time in the sleeping center might be
caused by environmental change and should have another
PSG examination to confirm OSA diagnosis. By using ox-
imeter to test OSA at home may eliminate such variation.
With regard to the second possibility, the data col-
lected from the 76 (12.34%) subjects who had TST less
than 4 hours were removed, resulting in a total of 540
subjects. To compare the accuracy in the discrimination
of 4 different groups of subjects (Table 2), a greater in-
crease in detecting mild (62.99% vs 70.25%) patients can
be observed, while it is only small difference in accuracy
regarding detection of normal (73.61% vs 75.00%), mod-
erate (41.67% vs 46.49%), and severe (88.07% vs 87.76%)
groups. Moreover, as shown in Tables 3 and 4, there is
no significant difference in diagnostic performance when
diagnosing severe patients or moderate and severe pa-
tients. The effect of less sleeping time needs to be further
investigated.
Study limitations
Unlike PSG and single-lead ECG, the limitation of oxim-
etry measurement is that it is unable to score sleep quality
[45]. Standard PSG scores sleep quality based on EEG sig-
nal analysis by grading the sleep quality into 4 stages of
continuum of depth during non-REM sleep. Thomas and
colleagues reported that sleep could be identified as wake/
REM, cyclic alternating pattern (CAP), and non-CAP
based on the Fourier analysis of R-R interval series and its
associated ECG-derived respiration (EDR) signal [45]. The
main advantage of using a combination of oximetry pa-
rameters as the predictor is its great sensitivity (>90%) in
the diagnosis of severe OSA patients with cheaper price
compared to PSG.
It was reported that study subjects selected based on
whether they has been referred for the index test instead
of clinical symptoms tended to have lower accuracy,
whereas studies with nonconsecutive inclusion of sub-
jects, retrospective selection of data, and discrimination
between healthy subjects and severe patients had signifi-
cantly higher accuracy [46]. Although our data were
Table 6 Comparison of diagnostic performance for SVM
models with different kernels in the diagnosis of
moderate and severe patients with AHI = 30 as the
threshold using ODI features
Data Predictive index (%) Linear Polynomial RBF
Dataset 1 (N = 616) Accuracy 90.25 90.09 90.42
Sensitivity 87.36 90.17 87.36
Specificity 92.74 90.03 93.05
Dataset 2 (N = 540) Accuracy 89.62 89.25 90.55
Sensitivity 89.45 89.45 89.87
Specificity 89.76 89.10 91.08
Hang et al. BMC Pulmonary Medicine (2015) 15:24 Page 11 of 13
collected consecutively from Jan. 2004 to Dec. 2015 and
the moderate and severe OSA patients were discrimi-
nated with a clear threshold of AHI = 15 and AHI = 30,
respectively, expecting not to overestimate the diagnostic
accuracy, several limitations must be considered when
interpreting the findings. First, this is a tertiary referral
hospital and the subjects were recruited from the outpa-
tients with suspicious OSA. Among the 616 subjects
studied, only 72 were verified as normal accounting to
only 11.69% of the total subjects recruited, which was
much less than the moderate (21.43%) and severe
(46.27%) patients. Moreover, the study subjects selected
based on the referred PSG examinations instead of clin-
ical symptoms tended to underestimate the diagnostic
accuracy [46]. Second, all events found in PSG were
verified by the technicians working in the sleep center.
The variation occurred among different technicians can-
not be avoided and neglected since each one has his/her
subjective judgment. Therefore, designing an objective
computer-assisted system to eliminate the loads and
subjective opinions of individual technicians is war-
ranted. Third, the oximetry is part of the gold standard
PSG examination, which reduces the quality of the study
and can artificially increase the diagnostic accuracy using
the parameters derived from the measurement of home-
style oximetry for its liability to induce errors and to
couple noise artifacts [47]. Finally, due to the design of this
study limiting the applicability of the results at home, valid
conclusions cannot be drawn about the accuracy of the
oximetry to detect or exclude OSA patients outside the
sleep laboratory.
Conclusion
Overnight pulse oximetry provides satisfactory diagnostic
performance in detecting severe OSA patients. Home-
styled oximetry may provide a tool for the diagnosis of
severe OSA patients. In addition to ODI parameters
(ODI2 and ODI4A), other variables derived from the
oximetry, such as heart rate variability (HRV) may also
be considered to be used for designing computer-
assisted diagnostic system to improve the diagnostic
performance.
Competing interests
The authors declare that they do not have competing interests.
Authorscontributions
These authorsindividual contributions were as follows. Conception and
design: LWH, JHC, WSC, and YFC. Administrative support: LWH, HLW, and
HHL. Collection and assembly of data: All authors. Data analysis and
interpretation: WSC and YFC. Manuscript writing: All authors. Final approval
of manuscript: All authors.
Acknowledgement
This study was supported in part by Ministry of Science and Technology,
Taiwan under grant no. NSC98-2410-H-039-003-MY2 and Taichung Hospital,
Ministry of Health and Welfare under grant No. FL1030505007-2.
Author details
1
Sleep Medicine Center, Department of Internal Medicine, China Medical
University Hospital, Taichung, Taiwan.
2
Department of Respiratory Therapy,
College of Health Care, China Medical University, Taichung, Taiwan.
3
Department of Beauty Science, National Taichung University of Science and
Technology, Taichung, Taiwan.
4
Department of Health Services
Administration, China Medical University, Taichung, Taiwan.
5
Department of
Internal Medicine, Taipei Hospital, Ministry of Health and Welfare, New Taipei
City, Taiwan.
6
Department of Management Information System, Central
Taiwan University of Science and Technology, Taichung, Taiwan.
7
Department of Internal Medicine, Taichung Hospital, Ministry of Health and
Welfare, Taichung, Taiwan.
8
Department of Healthcare Administration,
Central Taiwan University of Science and Technology, Taichung, Taiwan.
9
Department of Dental Technology and Materials Science, Central Taiwan
University of Science and Technology, Taichung, Taiwan.
Received: 21 September 2013 Accepted: 4 March 2015
References
1. Pang KP, Terris DJ, Podolsky R. Severity of obstructive sleep apnea:
correlation with clinical examination and patient perception. Otolaryngol
Head Neck Surg. 2006;135:55560.
2. Somers VK, White DP, Amin R, Abraham WT, Costa F, Culebras A, et al. Sleep
apnea and cardiovascular disease: an american heart association/american
college of cardiology foundation scientific statement from the american heart
association council for high blood pressure research professional education
committee, council on clinical cardiology, stroke council, and council on
cardiovascular nursing. In collaboration with the national heart, lung, and
blood institute national center on sleep disorders research (national institutes
of health). Circulation. 2008;118:1080111.
3. Kasai T, Floras JS, Bradley TD. Sleep apnea and cardiovascular disease: a
bidirectional relationship. Circulation. 2012;126:1495510.
4. Young T, Palta M, Dempsey J, Skatrud J, Weber S, Badr S. The occurrence of
sleep-disordered breathing among middle-aged adults. N Engl J Med.
1993;328:12305.
5. Young T, Evans L, Finn L, Palta M. Estimation of the clinically diagnosed
proportion of sleep apnea syndrome in middle-aged men and women.
Sleep. 1997;20:7056.
6. Hartenbaum N, Collop N, Rosen IM, Phillips B, George CF, Rowley JA, et al.
Sleep apnea and commercial motor vehicle operators: statement from the
joint task force of the american college of chest physicians, american
college of occupational and environmental medicine, and the national
sleep foundation. J Occup Environ Med. 2006;48:S437.
7. Lam JC, Lam B, Lam CL, Fong D, Wang JK, Tse HF, et al. Obstructive sleep
apnea and the metabolic syndrome in community-based Chinese adults in
Hong Kong. Respir Med. 2006;100:9807.
8. Lal C, Strange C, Bachman D. Neurocognitive impairment in obstructive
sleep apnea. Chest. 2012;141:160110.
9. Gosselin N, Mathieu A, Mazza S, Petit D, Malo J, Montplaisir J. Attentional
deficits in patients with obstructive sleep apnea syndrome: an event-related
potential study. Clin Neurophysiol. 2006;117:222835.
10. Teloken PE, Smith EB, Lodowsky C, Freedom T, Mulhall JP. Defining
association between sleep apnea syndrome and erectile dysfunction.
Urology. 2006;67:10337.
11. Pang KP, Terris DJ. Screening for obstructive sleep apnea: an evidence-based
analysis. Am J Otolaryngol. 2006;27:1128.
12. Flemons WW, Littner MR, Rowley JA, Gay P, Anderson WM, Hudgel DW,
et al. Home diagnosis of sleep apnea: a systematic review of the literature.
An evidence review cosponsored by the american academy of sleep
medicine, the american college of chest physicians, and the american
thoracic society. Chest. 2003;124:154379.
13. Netzer N, Eliasson AH, Netzer C, Kristo DA. Overnight pulse oximetry for
sleep-disordered breathing in adults: a review. Chest. 2001;120:62533.
14. Choi S, Bennett LS, Mullins R, Davies RJ, Stradling JR. Which derivative from
overnight oximetry best predicts symptomatic response to nasal continuous
positive airway pressure in patients with obstructive sleep apnoea? Respir
Med. 2000;94:8959.
15. Teramoto S, Matsuse T, Fukuchi Y. Clinical significance of nocturnal oximeter
monitoring for detection of sleep apnea syndrome in the elderly. Sleep
Med. 2002;3:6771.
Hang et al. BMC Pulmonary Medicine (2015) 15:24 Page 12 of 13
16. Vazquez J, Tsai W, Flemons W, Masuda A, Brant R, Hajduk E, et al.
Automated analysis of digital oximetry in the diagnosis of obstructive sleep
apnoea. Thorax. 2000;55:3027.
17. Kirk VG, Bohn SG, Flemons WW, Remmers JE. Comparison of home oximetry
monitoring with laboratory polysomnography in children. Chest.
2003;124:17028.
18. Lewis CA, Eaton TE, Fergusson W, Whyte KF, Garrett JE, Kolbe J. Home
overnight pulse oximetry in patients with COPD: more than one recording
may be needed. Chest. 2003;123:112733.
19. Brouillette RT, Morielli A, Leimanis A, Waters KA, Luciano R, Ducharme FM.
Nocturnal pulse oximetry as an abbreviated testing modality for pediatric
obstructive sleep apnea. Pediatrics. 2000;105:40512.
20. Zheng H, Sowers M, Buysse DJ, Consens F, Kravitz HM, Matthews KA, et al.
Sources of variability in epidemiological studies of sleep using repeated
nights of in-home
polysomnography: SWAN sleep study. J Clin Sleep Med JCSM Off Publ Am
Academy Sleep Med. 2012;8:8796.
21. Ahmadi N, Shapiro GK, Chung SA, Shapiro CM. Clinical diagnosis of sleep
apnea based on single night of polysomnography vs. two nights of
polysomnography. Sleep Breath. 2009;13:2216.
22. Newell J, Mairesse O, Verbanck P, Neu D. Is a one-night stay in the lab really
enough to conclude? First-night effect and night-to-night variability in
polysomnographic recordings among different clinical population samples.
Psychiatry Res. 2012;200:795801.
23. Al-Angari HM, Sahakian AV. Automated recognition of obstructive
sleep apnea syndrome using support vector machine classifier. IEEE
Transactions Information Technol Biomed Publ IEEE Engineer Med Biol
Soc. 2012;16:4638.
24. Marcos JV, Hornero R, Alvarez D, del Campo F, Zamarron C. Assessment of
four statistical pattern recognition techniques to assist in obstructive sleep
apnoea diagnosis from nocturnal oximetry. Med Eng Phys. 2009;31:9718.
25. Magalang UJ, Dmochowski J, Veeramachaneni S, Draw A, Mador MJ, El-Solh
A, et al. Prediction of the apnea-hypopnea index from overnight pulse
oximetry. Chest. 2003;124:1694701.
26. Berry RB, Budhiraja R, Gottlieb DJ, Gozal D, Iber C, Kapur VK, et al. Rules for
scoring respiratory events in sleep: update of the 2007 AASM manual for
the scoring of sleep and associated events. Deliberations of the sleep apnea
definitions task force of the american academy of sleep medicine. J Clin
Sleep Med JCSM Off Publ Am Academy Sleep Med. 2012;8:597619.
27. Iber C, Ancoli-Israel S, Chesson ALJ, Quan SF. The AASM manual for the
scoring sleep and associated events: rules, terminology and technical
specifications, 1
st
ed. Westchester, Illinois: American Academy of Sleep
Medicine; 2007.
28. Gyulay S, Olson LG, Hensley MJ, King MT, Allen KM, Saunders NA. A
comparison of clinical assessment and home oximetry in the diagnosis of
obstructive sleep apnea. Am Rev Respir Dis. 1993;147:503.
29. Brown MP, Grundy WN, Lin D, Cristianini N, Sugnet CW, Furey TS, et al.
Knowledge-based analysis of microarray gene expression data using sup-
port vector machines. PNAS. 2000;97:2627.
30. DeCoste D, Schuolkopf B. Training invariant support vector machines. Mach
Learn. 2002;46:16190.
31. LeCun Y, Jackel L, Bottou L, Brunot A, Cortes C, Denker J, et al. Comparison
of learning algorithms for handwritten digit recognition. In International
Conference on Artificial Neural Networks. Fogelman F, Gallinari P, editors.
1995:5360.
32. Lin HH, Hsu JC, Hsu YN, Pan RH, Chen YF, Tsang LY. Disulfide connectivity
prediction based on structural information without a prior knowledge of
the bonding state of cysteines. Computers in Biology and Medicine.
2013;43:19418.
33. Zhang J, Wang Y, Dong Y, Wang Y. Computer-aided diagnosis of cervical
lymph nodes on ultrasonography. Comput Biol Med. 2008;38:23443.
34. Zhi L, Zhang D, Yan J, Li Q, Tang Q. Classification of hyperspectral medical
tongue images for tongue diagnosis. Comput Med Imaging Graph.
2007;31:6728.
35. Eom JH, Kim SC, Zhang BT. AptaCDSS-E: A classifier ensemble-based clinical
decision support system for cardiovascular disease level prediction. Expert
Systems Appl. 2008;34:246579.
36. Polat K, Gunes S. Breast cancer diagnosis using least square support vector
machine. Digital Signal Process. 2007;17:694701.
37. Hsu JC, Chen YF, Chung WS, Tan TH, Chen T, Chiang JY. Clinical verification
of a clinical decision support system for ventilator weaning. Biomed Eng
Online. 2013;12 Suppl 1:S4.
38. Goncalves MA, Paiva T, Ramos E, Guilleminault C. Obstructive sleep apnea
syndrome, sleepiness, and quality of life. Chest. 2004;125:20916.
39. Dixon JB, Schachter LM, OBrien PE. Predicting sleep apnea and excessive
day sleepiness in the severely obese: indicators for polysomnography.
Chest. 2003;123:113441.
40. Khoo SM, Tan WC, Ng TP, Ho CH. Risk factors associated with habitual
snoring and sleep-disordered breathing in a multi-ethnic Asian population:
a population-based study. Respir Med. 2004;98:55766.
41. Lin CL, Yeh C, Yen CW, Hsu WH, Hang LW. Comparison of the indices of
oxyhemoglobin saturation by pulse oximetry in obstructive sleep apnea
hypopnea syndrome. Chest. 2009;135:8693.
42. Chen YF, Hang LW, Huang CS, Liang SJ, Chung WS. Polysomnographic
predictors of persistent continuous positive airway pressure adherence in
patients with moderate and severe obstructive sleep apnea. Kaohsiung J
Med Sci. 2015;31:839.
43. Schlotthauer G, Di Persia LE, Larrateguy LD, Milone DH. Screening of
obstructive sleep apnea with empirical mode decomposition of pulse
oximetry. Med Eng Phys. 2014;36:107480.
44. Doi Y, Minowa M, Tango T. Impact and correlates of poor sleep quality in
Japanese white-collar employees. Sleep. 2003;26:46771.
45. Thomas RJ, Mietus JE, Peng CK, Goldberger AL. An electrocardiogram-based
technique to assess cardiopulmonary coupling during sleep. Sleep.
2005;28:115161.
46. Rutjes AW, Reitsma JB, Di Nisio M, Smidt N, van Rijn JC, Bossuyt PM.
Evidence of bias and variation in diagnostic accuracy studies. CMAJ
Canadian Med Assoc J J de lAssoc Medicale Canadienne. 2006;174:46976.
47. Abd Sukor J, Mohktar MS, Redmond SJ, Lovell NH. Signal quality measures
on pulse oximetry and blood pressure signals acquired from self-measurement
in a home environment. IEEE J Biomed Health Inform. 2015;19:1028.
Submit your next manuscript to BioMed Central
and take full advantage of:
Convenient online submission
Thorough peer review
No space constraints or color figure charges
Immediate publication on acceptance
Inclusion in PubMed, CAS, Scopus and Google Scholar
Research which is freely available for redistribution
Submit your manuscript at
www.biomedcentral.com/submit
Hang et al. BMC Pulmonary Medicine (2015) 15:24 Page 13 of 13
... Over the last decade, there has been a considerable increase in interest and research activity in wearable sensors to measure blood oxygen saturation (SpO 2 ) [1][2][3][4][5][6][7][8][9]. ...
... Other approaches have suggested the use of 4G, WiFi, or Zigbee networks to upload data to a server, which could be accessed by remote coaches or doctors [4,8]. During a validation study against polysomnography in a sleep center, overnight pulse oximetry in a home setting was found to provide a satisfactory diagnostic performance in detecting severe obstructive sleep apnea [5]. The principle of pulse oximetry is typically applied, which assesses the level of blood oxygenation by optical light transmission through the blood (for a review of oximetry techniques and their theory, see Nitzan et al. [9]). ...
Article
Full-text available
This study reports the case of a lung cancer patient with increasing difficulties in falling asleep and frequent periods of wakefulness. Severe dyspnea related to pneumonitis caused as a side effect of immunotherapy worsened the situation. Eventually, a fear of falling asleep developed, including panic attacks and anxiety around choking, which was shown to lead to nights of complete wakefulness. The patient did not only sleep poorly; he did not sleep at all at night for several days, as evidenced by the notes he made during the night. Polygraphy showed no evidence of sleep-disordered breathing, but frequent periods of wakefulness and a reduced basal saturation of around 90% during sleep due to lung changes such as an extensive functional failure of the left upper lobe with position-dependent shunts. The authors hypothesized that the symptoms described were causally related to a drop in oxygen saturation in the patient’s blood. Therefore, they pursued the goal of finding a measurement technique that is as inexpensive as possible and that the patient can operate without outside assistance and great effort. Thus, the patient started using a low-cost wearable device that allows simultaneous measurements of blood oxygen content, pulse rate, and movement intensity. It consists of a finger ring with a pulse oximetry sensor and a wristband with a control unit containing a vibration motor. The described device reliably warned of disturbances in the oxygen concentration in the blood during the night with its vibration alarm. By use of that device during the whole night at home, the events of reduced oxygen saturation and anxiety symptoms were reduced. Sleep disturbances with sudden awakenings did not occur when using the device. The patient benefited from the security gained in this way and slept much more peacefully, and he could spend nights without waking up again. In conclusion, wearable oximeters with vibration alarms can be recommended for patients’ home care in lung cancer patients.
... Although several trials and studies have tried to simplify the detection of SA, few have focused on predicting SAEs. Several methods have highlighted the utility of the SpO 2 sensor for detecting OSA events [15,[19][20][21][22][23], but the American Academy of Sleep Medicine does not recommend using only this signal for detection [24] because sometimes SAEs occur without any notable change in oxygen saturation. Therefore, methods using SpO 2 must include another sensor to ensure a more reliable model. ...
Preprint
Full-text available
Sleep apnea (SA) is a common respiratory disorder, especially among obese people. It is caused by either the relaxation of the upper respiratory tract muscles or the failure of the neural signal to reach the muscles responsible for breathing, both of which interrupt the patient’s sleep–wake cycles. The traditional method for diagnosing this disorder, based on polysomnography, is complicated, vexing, expensive, time-consuming, and requires both sleep centers and specialized staff capable of connecting electrodes to the patient’s body. This paper proposes an SA prediction system based on merging five soft computing algorithms, specifically, combining the multi-verse optimizer (MVO) with an artificial neural network (ANN) to leverage measurements from heart rate, SpO2, and chest movement sensors. The most substantial novelty of this research is the hybridization of MVO and ANN (MVO-ANN), which improves the …
... PPG also has excellent value for detecting emergencies [95]. It is used nowadays mostly in emergency departments and in-home event detection [96], e.g., for overnight measurements [97]. ...
Article
Full-text available
The rapid development of elderly population is changing demographics in Europe and North America and imposes barriers to healthcare systems that may reduce the quality of service. Telemedicine is a potential solution supporting the real-time and remote monitoring of subjects as well as bidirectional communication with medical personnel for care delivery at the point of perception. Smart homes are private spaces where young or elderly, healthy or diseased-suffering, or disabled individuals spend the majority of their time. Hence, turning smart homes into diagnostic spaces for continuous, real-time, and unobtrusive health monitoring allows disease prediction and prevention before the subject perceives any symptoms. According to the World Health Organization, health, well-being, and quality of life assessment require the monitoring of interwoven domains such as environmental, behavioral, physiological, and psychological. In this work, we give an overview on sensing devices and technologies utilized in smart homes, which can turn the home into a diagnostic space. We consider the integration of sensing devices from all four WHO domains with respect to raw and processed data, transmission, and synchronization. We apply the bus-based scalable intelligent system to construct a hybrid topology for hierarchical multi-layer data fusion. This enables event detection and alerting for short-time as well as prediction and prevention for long-time monitoring.
... Although several trials and studies have tried to simplify the detection of SA, few have focused on predicting SAEs. Several methods have highlighted the utility of the SpO 2 sensor for detecting OSA events [15,[19][20][21][22][23], but the American Academy of Sleep Medicine does not recommend using only this signal for detection [24] because sometimes SAEs occur without any notable change in oxygen saturation. Therefore, methods using SpO 2 must include another sensor to ensure a more reliable model. ...
Article
Abstract Sleep apnea (SA) is a common respiratory disorder, especially among obese people. It is caused by either the relaxation of the upper respiratory tract muscles or the failure of the neural signal to reach the muscles responsible for breathing, both of which interrupt the patient’s sleep–wake cycles. The traditional method for diagnosing this disorder, based on polysomnography, is complicated, vexing, expensive, time-consuming, and requires both sleep centers and specialized staff capable of connecting electrodes to the patient’s body. This paper proposes an SA prediction system based on merging five soft computing algorithms, specifically, combining the multi-verse optimizer (MVO) with an artificial neural network (ANN) to leverage measurements from heart rate, SpO2, and chest movement sensors. The most substantial novelty of this research is the hybridization of MVO and ANN (MVO-ANN), which improves the ANN performance by selecting the best learning rate and number of neurons in hidden ANN layers. This enables highly accurate prediction of sleep apnea events. This work’s experimental results reveal that the MVO-ANN performs better than other algorithms, with mean absolute errors of 0.042, 0.202, and 0.166 for training, testing, and validation of the ANN. In addition, the SA prediction system achieved an accuracy of 98.67%, a sensitivity of 96.71%, and a specificity of 99.24%. These results provide good evidence that the proposed method can reliably predict respiratory events in people suffering from SA. Keywords Artificial neural network � Heart rate � Prediction � SpO2 � Sleep apnea � Soft computing algorithm
Book
Full-text available
(A&EI) as a novel subfield became obvious. As in all areas of Biomedical Informatics, A&EI must deal with issues such as relevant data collection, the management of data extracted from accident sites, health records or sensors, wearables and apps, and appropriate data processing, with the dual purpose of preventing harm and decision support. This book is an introduction to the research and application domain of A&EI, and is the product of three years’ work by the Working Group in A&EI of the International Medical Informatics Association (IMIA). The book presents ten chapters organized in four sections. The first section explores the framework for achieving an emergency-informatics health information infrastructure; the second focuses on the gathering of critical clinical data related to the building up of a smart environment for A&EI; the third introduces state-of-the-art technologies for integration into virtual emergency registries; and the final part considers the delicate issues of patient safety raised by the introduction of surveillance technologies into clinical care, along with other issues presenting challenges to the domain of A&EI for the future. The book is an important contribution to the field of A&EI, and will be of interest to healthcare professionals, informaticians, and all those who want a better understanding of the domain of Accident and Emergency Informatics.
Chapter
The 21st century has seen an enormous growth in emergency medical services (EMS) information technology systems, with corresponding accumulation of large volumes of data. Despite this growth, integration efforts between EMS-based systems and electronic health records, and public-sector databases have been limited due to inconsistent data structure, data missingness, and policy and regulatory obstacles. Efforts to integrate EMS systems have benefited from the evolving science of entity resolution and record linkage. In this chapter, we present the history and fundamentals of record linkage techniques, an overview of past uses of this technology in EMS, and a look into the future of record linkage techniques for integrating EMS data systems including the use of machine learning-based techniques.
Article
Background Skeletal nongrowing class II patients with retrognathic mandible are often at risk of developing obstructive sleep apnea (OSA). This skeletal discrepancy, when severe enough, is often addressed by surgical mandibular advancement which not only achieves aesthetics, structural balance, and functional efficiency but also improves the airway function. Objective The aim of the study was to evaluate the effectiveness of surgical mandibular advancement on airway dimensions, in nongrowing skeletal class II patients with mandibular retrognathism associated with OSA. Materials and Methods The sample consisted of 8 patients with skeletal class II due to retrognathic mandible associated with OSA. Screening was done with functional outcomes of sleep questionnaire (FOSQ) score and confirmed with polysomnography. Pretreatment and posttreatment airway dimensions/volume were assessed and calculated on cone beam computed tomography (CBCT). Three months post surgery, oxygen saturation was recorded. Results A statistically significant improvement in mean oxygen saturation level was observed from 87.62 ± 7.70 to 97.75 ± 0.46 and mean increase in airway volume was from 48.60 ± 7.17 to 69.48 ± 26.22. The smallest cross section increased from 137 ± 74.37 to 253.75 ± 76.67. Anteroposterior dimensions increased significantly from 6.17 ± 1.78 to 9.33 ± 2.07 and transversed from 24.56 ± 6.41 to 30.14 ± 3.35. FOSQ score improved from 9.37 ± 2.61 to 15.25 ± 0.88. Conclusion Surgical mandibular advancement is an effective treatment option in improving airway dimensions in nongrowing skeletal class II patients with mandibular retrognathism associated with OSA.
Article
Full-text available
To date, although many studies had focused on the impact of environmental factors on sleep, how to choose the proper assessment method for objective sleep quality was often ignored, especially for healthy subjects in bedroom environment. In order to provide methodological guidance for future research, this paper reviewed the assessments of objective sleep quality applied in environmental researches, compared them from the perspective of accuracy and interference, and statistically analyzed the impact of experimental type and subjects’ information on method selection. The review results showed that, in contrast to polysomnography (PSG), the accuracy of actigraphy (ACT), respiratory monitoring-oxygen saturation monitoring (RM-OSM), and electrocardiograph (ECG) could reach up to 97, 80.38, and 79.95%, respectively. In terms of sleep staging, PSG and ECG performed the best, ACT the second, and RM-OSM the worst; as compared to single methods, mix methods were more accurate and better at sleep staging. PSG interfered with sleep a great deal, while ECG and ACT could be non-contact, and thus the least interference with sleep was present. The type of experiment significantly influenced the choice of assessment method (p < 0.001), 85.3% of researchers chose PSG in laboratory study while 82.5% ACT in field study; moreover, PSG was often used in a relatively small number of young subjects, while ACT had a wide applicable population. In general, researchers need to pay more attention at selection of assessments in future studies and this review can be used as a reliable reference for experimental design.
Chapter
Sleeping disorder can affect many facets of life and can be common health problems. The most prevalent sleep disorder is Sleep Apnea (SA) which is referred to as a decreased or a loss of ventilation in the lungs in sleep. This paper suggests an alternate machine-based system of clinical Polysomnography (PSG), a multimodal, feature-based computing pipeline for the detection of Obstructive Sleep Apnea (OSA). The procedure suggested includes the fusion at the classification stage of Oxygen Saturation (SpO2) and the signals Electrocardiogram (ECG). The Kernel Principal Component Analysis (KPCA) feature selection approach is used to achieve robust features of all signal sources and to reduce the dimensionality of features. Using the Support Vector Machine (SVM) and Random Forest classifiers, the efficiency of the feature selection techniques used was analyzed.
Article
Full-text available
Extensive use of continuous positive airway pressure (CPAP) has positive clinical benefits for most patients with obstructive sleep apnea (OSA). However, patient adherence is a major limiting factor to the effectiveness of CPAP treatment. This study determined the potential and quantifiable factors affecting the willingness of patients with OSA to undertake CPAP treatment by comparing the polysomnographic parameters recorded during diagnosis and titration. Patients with moderate and severe OSA who attended diagnostic polysomnography (PSG) and CPAP titration at the sleep center of China Medical University Hospital (CMUH) were included in the study. A total of 312 patients were divided into persistent users and nonusers of CPAP according to their use of in-home CPAP following titration and a 7-day CPAP trial. Multivariate logistic regression analyses were used to define the potential polysomnographic predictors of persistent CPAP adherence, and odds ratios (ORs) and 95% confidence intervals (CIs) were calculated. Most patients were men older than 50 years who were overweight or obese. Among the patients, 146 (46.8%) became persistent CPAP users. A 10% improvement of oxygen desaturation index (ODI) and a 10% increment in deep sleep percentage increased the chance of persistent CPAP use 1.18-fold and 1.07-fold, respectively. In addition, the improved ODI and deep sleep during CPAP titration increased the chance of persistent CPAP user. The polysomnographic parameters obtained from diagnosis and during titration can facilitate the prediction of persistent CPAP use.
Article
Full-text available
Weaning is typically regarded as a process of discontinuing mechanical ventilation in the daily practice of an intensive care unit (ICU). Among the ICU patients, 39%-40% need mechanical ventilator for sustaining their lives. The predictive rate of successful weaning achieved only 35-60% for decisions made by physicians. Clinical decision support systems (CDSSs) are promising in enhancing diagnostic performance and improve healthcare quality in clinical setting. To our knowledge, a prospective study has never been conducted to verify the effectiveness of the CDSS in ventilator weaning before. In this study, the CDSS capable of predicting weaning outcome and reducing duration of ventilator support for patients has been verified. A total of 380 patients admitted to the respiratory care center of the hospital were randomly assigned to either control or study group. In the control group, patients were weaned with traditional weaning method, while in the study group, patients were weaned with CDSS monitored by physicians. After excluding the patients who transferred to other hospitals, refused further treatments, or expired the admission period, data of 168 and 144 patients in the study and control groups, respectively, were used for analysis. The results show that a sensitivity of 87.7% has been achieved, which is significantly higher (p<0.01) than the weaning determined by physicians (sensitivity: 61.4%). Furthermore, the days using mechanical ventilator for the study group (38.41 ± 3.35) is significantly (p<0.001) shorter than the control group (43.69 ± 14.89), with a decrease of 5.2 days in average, resulting in a saving of healthcare cost of NT$45,000 (US$1,500) per patient in the current Taiwanese National Health Insurance setting. The CDSS is demonstrated to be effective in identifying the earliest time of ventilator weaning for patients to resume and sustain spontaneous breathing, thereby avoiding unnecessary prolonged ventilator use and decreasing healthcare cost.
Article
Background Obstructive sleep apnea (OSA) is common in severely obese subjects (body mass index [BMI] > 35). Overnight polysomnography (OPS) is the “gold standard” method of evaluating this condition; however, it is time-consuming, inconvenient, and expensive. Selection of patients for OPS would be enhanced if we could better predict those likely to have clinically significant OSA. Study objective To look for clinical and biochemical predictors of OSA in symptomatic patients presenting for obesity surgery. Design and patients Symptoms suggestive of OSA were sought in a structured interview. We report OPS results of 99 consecutive subjects in whom OSA was clinically suspected. Predictors of apnea-hypopnea index (AHI) were sought from an extensive preoperative data collection. Multivariate linear and logistic analysis was used to identify independent predictors of AHI. Results Symptoms were poor predictors of AHI, with observed sleep apnea the only positive predictor. Four clinical and two biochemical factors independently predicted AHI: observed sleep apnea, male sex, higher BMI, age, fasting insulin, and glycosylated hemoglobin AIc (r² = 0.42). Neck circumference (the best single measure) could replace BMI and sex in the analysis (r² = 0.43). With cutoffs selected, a simple scoring system using these six factors provides a method of predicting those with moderate or severe OSA. A score ≥ 3 provides a sensitivity and specificity of 89% and 81%, and 96% and 71% for AHIs of ≥ 15 and ≥ 30, respectively. None of the 31 subjects with scores of 0 or 1 were found to have an AHI ≥ 15. Conclusion We explore sleep disturbance and report a simple method of predicting OSA in severely obese symptomatic subjects. This should assist in limiting the use of OPS to those with greater risk and provide a method of assessing risk in those not presenting primarily with a sleep problem.
Article
Study objectives To measure the accuracy and reliability of a portable home oximetry monitor with an automated analysis for the diagnosis of obstructive sleep apnea (OSA) in children. Design Prospective cohort study. Setting Alberta Lung Association Sleep Center, Alberta Children’s Hospital Sleep Clinic. Study subjects Consecutive, otherwise healthy children, aged 4 to 18 years, presenting to the Pediatric Sleep Service at the Alberta Children’s Hospital for assessment of possible OSA. Interventions All subjects underwent 2 nights of monitoring in the home with an oximetry-based portable monitor with an automatic internal scoring algorithm. A third night of monitoring was done simultaneously with computerized laboratory polysomnography according to American Thoracic Society guidelines. Measurements and results Both test-retest reliability of the portable monitor-based desaturation index (DI) between 2 nights at home and between laboratory and home were high using the Bland and Altman analysis (mean agreement, 0.32 and 0.64; limits of agreement, − 8.00 to 8.64 and − 0.75 to 6.50, respectively). The polysomnographic apnea-hypopnea index (AHI) agreed poorly with the portable monitor DI (mean difference, 1.27; limits of agreement, − 12.02 to 15.02). The sensitivity and specificity of the monitor for the identification of moderate sleep apnea (polysomnography AHI > 5/h) were 67% and 60%, respectively. Conclusion Portable monitoring based only on oximetry alone is not adequate for the identification of OSA in otherwise healthy children.
Article
Recently, decision support system (DSSs) have become more widely accepted as a support tool for use with telehealth systems, helping clinicians to summarise and digest what would otherwise be an unmanageable volume of data. One of the pillars of a home telehealth system is the performance of unsupervised physiological self-measurement by patients in their own homes. Such measurements are prone to error and noise artifact, often due to poor measurement technique and ignorance of the measurement and transduction principles at work. These errors can degrade the quality of the recorded signals and ultimately degrade the performance of the DSS system which is aiding the clinician in their management of the patient. Developed algorithms for automated quality assessment for pulse oximetry and blood pressure (BP) signals were tested retrospectively with data acquired from a trial that recorded signals in a home environment. The trial involved four aged subjects who performed pulse oximetry and BP measurements by themselves at their home for ten days, three times per day. This trial was set up to mimic the unsupervised physiological self-measurement as in a telehealth system. A manually annotated 'gold standard' (GS) was used as the reference against which the developed algorithms were evaluated after analysing the recordings. The assessment of pulse oximetry signals shows 95% of good sections and 67% of noisy sections were correctly detected by the developed algorithm, and a Cohen's Kappa coefficient (κ) of 0.58 was obtained in 120 pooled signals. The BP measurement evaluation demonstrates that 75% of the actual noisy sections were correctly classified in 120 pooled signals, with 97% and 91% of the signals correctly identified as worthy of attempting systolic and/or diastolic pressure estimation, respectively, with a mean error and standard deviation of 2.53±4.20 mmHg and 1.46±5.29 mmHg when compared to a manually annotated GS. These results demonstrate the feasibility, and highlight the potential benefit, of incorporating automated signal quality assessment algorithms for pulse oximetry and BP recording within a DSS for telehealth patient management.
Article
Detection of desaturations on the pulse oximetry signal is of great importance for the diagnosis of sleep apneas. Using the counting of desaturations, an index can be built to help in the diagnosis of severe cases of obstructive sleep apnea-hypopnea syndrome. It is important to have automatic detection methods that allows the screening for this syndrome, reducing the need of the expensive polysomnography based studies. In this paper a novel recognition method based on the empirical mode decomposition of the pulse oximetry signal is proposed. The desaturations produce a very specific wave pattern that is extracted in the modes of the decomposition. Using this information, a detector based on properly selected thresholds and a set of simple rules is built. The oxygen desaturation index constructed from these detections produces a detector for obstructive sleep apnea-hypopnea syndrome with high sensitivity ($0.838$) and specificity ($0.855$) and yields better results than standard desaturation detection approaches.