Received June 26, 2019, accepted July 1, 2019, date of publication July 9, 2019, date of current version July 29, 2019.
Digital Object Identifier 10.1109/ACCESS.2019.2927781
Personalized Maternal Sleep Quality Assessment:
An Objective IoT-based Longitudinal Study
IMAN AZIMI 1, (Student Member, IEEE), OLUGBENGA OTI1, SINA LABBAF2,
HANNAKAISA NIELA-VILÉN3, ANNA AXELIN3, NIKIL DUTT2, (Fellow, IEEE),
PASI LILJEBERG 1, (Member, IEEE), AND AMIR M. RAHMANI2,4, (Senior Member, IEEE)
1Department of Future Technologies, University of Turku, FI-20014 Turku, Finland
2Department of Computer Science, University of California at Irvine, Irvine, CA 92697, USA
3Department of Nursing Science, University of Turku, FI-20014 Turku, Finland
4School of Nursing, University of California at Irvine, Irvine, CA 92697, USA
Corresponding author: Iman Azimi (imaazi@utu.fi)
This work was supported in part by the Academy of Finland through the PREVENT Project under Grant 313448 and Grant 313449, in part
by the Academy of Finland through the SLIM Project under Grant 316810 and Grant 316811, and in part by the U.S. National Science
Foundation (NSF) through the UNITE Project under Grant SCC CNS-1831918.
ABSTRACT Sleep is a composite of physiological and behavioral processes that undergo substantial changes
during and after pregnancy. These changes might lead to sleep disorders and adverse pregnancy outcomes.
Several studies have investigated this issue; however, they were restricted to subjective measurements or
short-term actigraphy methods. This is insufficient for a longitudinal maternal sleep quality evaluation.
A longitudinal study: 1) requires a long-term data collection approach to acquire data from everyday routines
of mothers and 2) demands a sleep quality assessment method exploiting a large volume of multivariate data
to assess sleep adaptations and overall sleep quality. In this paper, we present an Internet-of-Things-based
long-term monitoring system to perform an objective sleep quality assessment. We conduct longitudinal
monitoring, where 20 pregnant mothers are remotely monitored for six months of pregnancy and one month
postpartum. To evaluate sleep quality adaptations, we: 1) extract several sleep attributes and study their
variations during the monitoring and 2) propose a semi-supervised machine learning approach to create
a personalized sleep model for each subject. The model provides an abnormality score, which allows an
explicit representation of the sleep quality in a clinical routine, reflecting possible sleep quality degradation
with respect to her own data. Sleep data of 13 participants (out of 20) are included in our analysis, as their data
are adequate for the study, including 172.15±33.29 days of sleep data per person. Our fine-grained objective
measurements indicate that the sleep duration and sleep efficiency are deteriorated in pregnancy and notably
in postpartum. In comparison to the mid of the second trimester, the sleep model indicates the increase of
sleep abnormality at the end of pregnancy (2.87 times) and postpartum (5.62 times). We also show that the
model enables individualized and effective care for sleep disturbances during pregnancy, as compared to a
baseline method.
INDEX TERMS Anomaly detection, Internet of Things, longitudinal study, maternity care, sleepmonitoring,
sleep quality assessment.
I. INTRODUCTION
Several physical, physiological, and hormonal adaptations
occur during pregnancy to accommodate the developing fetus
and to prepare the mother for the delivery [1], [2]. Such vari-
ations in the maternal body alter sleep patterns of pregnant
women in many ways. In this regard, sleep disturbances are
The associate editor coordinating the review of this manuscript and
approving it for publication was Mahmoud Barhamgi.
particularly prevalent throughout the pregnancy, including
various disorders to maintaining sleep (e.g., insomnia), sleep
deprivation, and restless legs syndrome [3]–[6]. Moreover,
sleep patterns of pregnant women might be altered in post-
partum months, as they experience new life situations after
labor [7].
Studies show that sleep disturbances negatively impact
maternal and child health during and after pregnancy [8].
Sleep problems are associated with a high likelihood of
VOLUME 7, 2019 This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see http://creativecommons.org/licenses/by/4.0/ 93433
I. Azimi et al.: Personalized Maternal Sleep Quality Assessment: An Objective IoT-based Longitudinal Study
poor obstetric outcomes and different diseases such as ges-
tational diabetes, preeclampsia, and stress overload [9]–[11].
Also, they lead to increased risk of preterm birth,
intrauterine growth restriction, and unplanned Caesarean
deliveries [12], [13]. Moreover, different studies discussed
the correlation between sleep disturbances and postpartum
diseases and complications such as depression and damage
to the mother-infant relationship [12], [14]. Thus, screening,
monitoring, and assessment of maternal sleep quality are
essential during pregnancy to alleviate sleep disturbances and
prevent its potential complications [12], [15], [16].
Sleep quality is a complex concept that is traditionally eval-
uated via qualitative attributes (i.e., subjectivemeasurements)
and more recently via quantitative attributes (i.e., objec-
tive measurements) [17]. Subjective techniques determine
perceived sleep quality by inquiring the individuals about
their sleep experiences such as sleep duration and distur-
bances. These techniques are often performed via self-report
questionnaires such as the Pittsburgh Sleep Quality Index
(PSQI) [18] and Berlin Questionnaire [19]. Those are widely
used in sleep quality evaluation of different groups of people
as they are relatively straightforward and easy to implement
for longitudinal studies. Similar subjective techniques have
also been utilized for pregnant women to reveal the impact
of pregnancy onmaternal sleep [5], [8], [15], [16], [20]–[22].
However, such subjective methods can be inaccurate and
poorly reflect sleep quality level, as the data collection is
mostly limited to scheduled interviews, Internet-based sur-
veys, or self-report questionnaires. The shortcomings and
poor performance of such methods have been widely dis-
cussed in several studies investigating the validity of the
subjective sleep quality assessment methods [17], [23]–[25].
Alternatively, objective techniques measure the user’s
physical and health conditions and translate the results into
sleep attributes such as sleep efficiency and sleep stages
for further assessment. Polysomnography (PSG) is a con-
ventional test in this regard, where several bio-signals are
acquired for sleep analysis [26], [27]. The PSG, as the gold
standard of the sleep assessment, has been exploited for sleep
disturbances monitoring in pregnancy [4]. However, it is
bounded to one or a limited number of nights due to its data
acquisition limits. Actigraphy is another objective method
that examines sleep quality bymonitoring human rest/activity
cycles [28]. Data acquisition in actigraphy is more conve-
nient and non-invasive for users, as it is performed via a
small and light-weight wearable device placed on the user’s
wrist or ankle. Standalone (i.e., without network connectivity
and real-time remote access) actigraphy monitors have been
deployed for offline and short-time sleep monitoring, such
as the works presented in [29]–[32] where maternal sleep is
monitored for up to 14 days. However, the constraints in local
storage and processing have hindered the utilization of this
technology for longitudinal sleep quality monitoring.
Longitudinal objective sleep monitoring necessitates a
long-term data collection to acquire data from everyday rou-
tines of participants 24/7. We believe recent advancements
in Internet-of-Things (IoT) technologies provide an unprece-
dented opportunity to enable such continuous health mon-
itoring. IoT is an emerging network of interrelated objects
that tailors a distinct set of paradigms such as wearable
electronics, communication infrastructure, and data analytics
to deliver personalized services to the end-users [33], [34].
However, it should be noted that an IoT-based sleep monitor-
ing system, despite being a powerful tool, generates a large
volume of multivariate data which dramatically increases
over time. Such big data [35], while being a rich source of
information, call for tailored and intelligent data analytic
techniques and models.
Conventional techniques assess the sleep quality only from
a single perspective by separately extracting and analyzing
each sleep attribute (e.g., sleep duration) from a pool of
sleep-related data. Data in a high-dimensional space require
a more intelligent amalgamation method to transform all
sleep attributes into a single overall sleep quality score,
in a way that the contribution of each attribute is automat-
ically considered in the final score. This allows a straight-
forward representation of the sleep quality in a clinical
routine and reflecting possible sleep quality degradation of
an individual with respect to her own life situation and
health condition. We believe that such a method is partic-
ularly essential for maternal sleep quality assessment and
individualized care approach, as a mother’s physical and
mental states undergo a process of change throughout the
course of pregnancy and postpartum, which necessitates an
explicit indicator of the mother’s sleep changes during this
period.
In this paper, we present an IoT-based long-term monitor-
ing system that employs a wrist-worn device to assess the
sleep of pregnant women during pregnancy and postpartum
thoroughly. Our monitoring system is deployed on a real
human subject trial where 20 pregnant women are remotely
and continuously monitored for six months of pregnancy and
one month postpartum. We first study sleep quality changes
in this monitoring, leveraging several objective attributes.
We then propose an anomaly detection approach to construct
a personalized sleep model for each individual using the
sleep data from the beginning of the monitoring process.
We measure the sleep adaptations of the rest of the pregnancy
and postpartum, using the personalized model to investigate
the maternal sleep quality from a different perspective. In
summary, the contribution of this paper is manifold:
i) Presenting an IoT-based long-term monitoring system
to perform objective sleep quality assessment during
pregnancy and postpartum.
ii) Conducting a longitudinal study on a human subject
trial on maternal sleep.
iii) Observing the degradation of sleep quality during
pregnancy and postpartum separately for a set of
fine-grained quantitative sleep attributes.
iv) Proposing a neural network-based approach to inves-
tigate maternal sleep quality adaptations in a compre-
hensive and personalized way.
93434 VOLUME 7, 2019
I. Azimi et al.: Personalized Maternal Sleep Quality Assessment: An Objective IoT-based Longitudinal Study
The rest of the paper is organized as follows.We outline the
background and related work of this research in Section II.
Section III describes the study design. In Section IV,
we present our sleep analysis approaches. Results and find-
ings are presented in Section V. In Section VI, we discuss our
findings, evaluate the model, and represent limitations and
future directions of this study. Finally, Section VII concludes
the paper.
II. BACKGROUND AND RELATED WORK
In this section, we first outline the background of maternal
health and sleepmonitoring. Then, we present state-of-the-art
anomaly detection techniques as appropriate tools to create
models for abnormality detection.
A. MATERNAL HEALTH AND SLEEP MONITORING
Maternal health can be monitored during pregnancy to ensure
the well-being of both the mother and her future child. Preg-
nancy is a window to a woman’s future health [36], and thus
women are also interested in monitoring their health during
pregnancy. Furthermore, using systematic and regular mon-
itoring, several abnormalities and complications regarding
pregnancy could be detected early and be treated accord-
ingly. Maternal health monitoring, however, varies in differ-
ent countries, and only half of women receive the recom-
mended amount of care during their pregnancy [37]. There-
fore, there is a need to develop new solutions that can widen
the availability of maternal health monitoring for all pregnant
women.
Sleep as an important part of overall maternal health
requires particular attention. Multiple hormonal and physi-
ological changes during pregnancy might contribute to sleep
problems. For example, nausea, vomiting, or anxiety might
cause sleep disturbances in the first trimester of pregnancy.
As pregnancy progresses, the frequency and duration of
sleep disturbances increase. Frequent urination, backache, leg
cramps, and anxiety about delivery are common reasons for
compromised sleep in the third trimester.
Sleep disturbances are common during pregnancy and
are the risk factors of adverse pregnancy outcomes such
as prenatal depression, gestational diabetes, and preterm
birth [11], [16], [38], [39]. Also, many women suffer from
acute sleep deprivation during the postpartum period, and
compromised sleep may continue even several months after
birth [39]. This problemmight lead to diseases such as mater-
nal fatigue and postpartum depression [14]. It is possible to
use nonpharmacological strategies such as regular physical
activity, controlling weight gain, and relaxation, to alleviate
sleep disorders during pregnancy. Medication should be used
only in severe cases to avoid possible teratogenic effects [40].
Sleep quality assessment is the first step for managing sleep
disturbances and disorders. It gives an accurate picture of
sleep changes and assists to early-detect sleep problems [41].
In particular, systematic and personalized sleep assessment
enables the provision of right strategies to manage sleep
disturbances and disorders of each woman.
Different methods have been proposed in the literature
to investigate sleep problems. The duration, as well as the
quality of sleep during pregnancy, has usually been measured
using questionnaires [16], [42], [43]. The Pittsburgh Sleep
Quality Index (PSQI) is the gold standard for subjective
sleep quality assessment, in which individuals are asked to
answer a self-report questionnaire [18]. The tool discrimi-
nates ‘‘good’’ sleep quality from ‘‘bad’’ leveraging seven
component scores such as sleep latency, habitual sleep effi-
ciency, and use of sleepingmedication. Such subjectivemeth-
ods are not accurate; pregnant women have both over and
underestimated their sleep duration compared with objective
measurements [44].
Polysomnography (PSG) is the gold standard of sleep
monitoring. The method typically employs various wearable
sensors to capture several bio-signals including electroen-
cephalogram (EEG), electromyogram (EMG), electroocu-
logram (EOG), and electrocardiogram (ECG), providing
different sleep indices such as sleep efficiency, sleep onset
latency, and sleep stages [4], [26], [45]. However, the use of
the PSG is limited to sleep laboratories and clinical settings
due to the burdensome implementation of its multisensor data
acquisition. Therefore, the method was mostly performed
in a short period of time in sleep studies. For example,
an overnight lab-based PSG was implemented along with the
Berlin questionnaire, targeting obstructive sleep apnea [19].
Similarly, in the maternity care, sleep disturbance was inves-
tigated via a short-term PSG-based data collection, i.e., two
consecutive nights in each trimester, and in first and third
postpartum months [4].
Actigraphy is another low-cost alternative for monitor-
ing sleep and sleep-wake behavior of an individual [28].
The Sleep actigraphy typically includes an actigraph device
equipped with a 3-axis MEMS accelerometer sensor, a low-
performance processor and a limited memory. The accel-
eration data are locally processed, and sleep parameters
are extracted. The actigraphy method is easy-to-use in
out-of-hospital settings in contrast to the PSG. However,
it is bounded to offline services. Objective sleep mon-
itoring has been fulfilled in different maternal studies
using short-term actigraphymethods [30], [46]. For example,
Lee and Gay [29] investigated the association between sleep
disturbance in late pregnancy with labor using an actigraphy
for 2 days along with subjective measurements in the ninth
month of pregnancy; a seven-day actigraphy and PSQI meth-
ods were employed for maternal sleep disturbance [31]; and
Haney et al. [32] assess sleep in early pregnancy exploiting
a 14-day actigraphy method, questionnaires, and blood pres-
sure measurements.
Contact-free sensors have also been proposed for sleep
monitoring. Some examples are visual-based sensors [47],
mattress-based sensors [48], and smartphone sensors [49].
They were mostly designed to acquire sleep patterns as well
as vital signs such as heart rate and respiration rate. The use
of such systems has been limited in real-world applications
because of restrictions in data collection and high cost. In one
VOLUME 7, 2019 93435
I. Azimi et al.: Personalized Maternal Sleep Quality Assessment: An Objective IoT-based Longitudinal Study
study, the maternal body movements of 2 pregnant women
were monitored for a couple of weeks, using a piezoelectric
sensor board placed beneath their mattress [50].
B. ANOMALY DETECTION
Anomaly detection, also known as outlier detection, is the
problem of finding patterns or events in data that differ
from the expected behavior [51]. Anomaly detection has been
applied in many fields including fraud detection, healthcare,
and intrusion detection in cybersecurity [52]. An anomaly
detection technique applied to a problem depends on a variety
of factors including the availability of labeled data, the nature
of the data, the type of anomalies to be detected, the output
of the method, and in some cases the field of study.
The type of anomalies in a dataset can be divided into
three major categories [51]. First, point anomalies refer to
data instances that are anomalous with respect to the rest
of the data (i.e., normal data). Second, contextual anomalies
are data instances that are anomalous in a certain context.
For example, 150 heart beats per minute would be normal
during exercise although it is anomalous if the user is sleep-
ing. Third, collective anomalies refer to a group of related
data instances which together are considered anomalous. For
instance, recording a couple of high heart rate events in a day
would be detected as anomalous (e.g., health deterioration)
in a health application. Moreover, datasets can be modified
to change the anomaly type; e.g., point anomalies and collec-
tive anomalies can become contextual anomalies if we add
context information to the dataset.
The choice of a specific anomaly detection method –
supervised, semi-supervised, and unsupervised – is greatly
dependent on the type of data involved. The data can gener-
ally be divided into binary, categorical, or continuous. How-
ever, it can be a combination of these categories in some
cases. In addition, the output of the method can be either
binary (i.e., normal or anomalous) or continuous in the form
of an anomaly score which represents the degree of the
anomaly [51]. The availability of labeled data is a common
challenge in anomaly detection, as anomalies might not occur
frequently. Moreover, labeling of a dataset by an expert is
time-consuming and expensive. The extent of the availability
of a labeled dataset determines which method is used.
Supervised anomaly detection methods rely on data with
labels for both the normal and anomalous classes. They con-
struct a predictivemodel to differentiate normal and abnormal
behavior. However, unbalanced distribution of data should be
considered in such models, as in practice anomalous data do
not occur as often as normal data. Examples of such meth-
ods include Neural Networks methods [53], Support Vector
Machine (SVM) [54], and Rule-based approaches (e.g. Deci-
sion trees) [55].
Semi-supervised anomaly detection methods deploy
semi-supervised learning (also known as one-class learn-
ing methods) that only consider normal data to train their
models. When the model is created to understand nor-
mal behavior, it can then distinguish between normal and
FIGURE 1. The IoT-based system for the maternal health monitoring.
anomalous classes. These methods are commonly applied
because of unavailability or shortage of anomalous data in
many applications. Moreover, no data labeling is required,
as all the input data are normal. Some examples of these
methods are Statistical techniques [56], one-class Sup-
port Vector Machine (SVM) [57], and Neural Networks
methods [58]–[60].
In contrast, unsupervised anomaly detection methods
deploy unsupervised learning techniques that require no train-
ing data, assuming the normal data occur more often than
anomalous data. Unfortunately, applying data that do not fit
this assumption would lead to a high false positive rate. Clus-
tering techniques [61] and Nearest Neighbor techniques [62]
are examples of unsupervised or semi-supervised techniques,
which rely on the assumption that normal data remain in a
cluster or dense neighborhood while anomalous data do not.
They often require large training data for the normal classes.
III. STUDY DESIGN
This paper proposes an IoT-based monitoring system
equipped with a semi-supervised machine learning approach,
by which pregnant women can be monitored remotely, con-
tinuously, and long-term. Also, the proposed system enables
personalized sleep analysis during pregnancy and the postpar-
tum, providing effective care for maternal sleep disturbances.
We present this system for a real human subject trial on
material sleep, where pregnant women are monitored in six
months of pregnancy and one month postpartum. In this
section, we introduce the IoT-based monitoring system and
provide details about our implementation setup, the partici-
pants, and recruitment.
A. IOT-BASED MONITORING SYSTEM
An IoT-based system is introduced to continuously monitor
the pregnant women. As shown in Figure 1, the architecture of
the proposed system is partitioned into three main tiers. First,
the sensor network performs data collection in IoT-based
systems, located in the vicinity of the end-users. It acquires
pregnancy- and sleep- related data from the end-users con-
stantly. Thanks to the advances in embedded and wearable
technologies, various lightweight energy-efficient wearable
devices such as smartwatches, fitness trackers and Holter
monitors are nowadays available for this tier.
93436 VOLUME 7, 2019
I. Azimi et al.: Personalized Maternal Sleep Quality Assessment: An Objective IoT-based Longitudinal Study
The gateway, as the second tier, is a bridge between the sen-
sor network and the Internet (i.e., cloud servers). The gateway
is responsible for data transmission and protocol conversion.
Smartphones and tablets as widespread mobile computing
devices can be employed in this layer. They provide data
transmission in both directions, transmitting collected health
data to the cloud servers as well as sending reports and
feedback to the end-user. Moreover, subjective measurements
including interviews and Internet-based surveys can be car-
ried out.
The cloud server, as the third tier, includes a high-
performance computing infrastructure. It is responsible for
the sleep quality analysis (e.g., data abstraction and model-
ing). Our semi-supervised machine learning approach is fully
positioned at this tier. Moreover, the cloud server manages,
secures, and stores the data remotely and is capable of pro-
viding a control panel for data visualization. The processed
data are shared with the experts (e.g., researchers) for further
analysis.
Setup: For the data collection, we restricted our selec-
tion of sensor nodes to wearable products (e.g., smart wrist-
bands and smartwatches) that are technically applicable and
practically feasible to continuous long-term monitoring [63].
Various studies have shown the validity and reliability of
such wearables in terms of sleep parameters by comparing
different wearables with the gold standard PSG [64]–[66].
At the beginning of the study, various devices such as Garmin
Vivosmart HR [67], Microsoft Band1 and Fitbit Charge HR2
were available in the market. We selected the Garmin Vivos-
mart HR considering several factors such as the built-in sen-
sors, battery life, small size, light weight, strap design, and
waterproofness. More details of the feasibility of this study
can be found in [68].
The Garmin Vivosmart HR contains an optical sensor and
an inertial measurement unit (IMU), through which photo-
plethysmogram (PPG) [69] and acceleration signals are col-
lected. In our setup, the participants were requested to wear
the device continuously. We acquired a set of data every
15 minutes, including heart rate, step counts, and body move-
ments. The data were utilized for the sleep analysis.
In addition, the pregnant women were asked to frequently
synchronize the wristband’s data with the remote servers via
gateway devices – their smartphones or personal computers
in this setup. For the server, we used a Linode virtual private
server (VPS) [70] with two 2.50GHz Intel Xeon CPU (E5-
2680 v3), 4GB memory, and SSD storage drive. The cloud
server was used to store the data remotely, to perform the
sleep quality analysis methods, and to provide data visual-
ization.
B. PARTICIPANTS AND RECRUITMENT
The monitoring was performed on primiparous preg-
nant women attending to one of two selected maternity
1https://support.microsoft.com/en-us/help/4000514/band-2-get-started
2https://www.fitbit.com/be/chargehr
TABLE 1. Background information of the selected participants.
outpatient clinics in Southern Finland BetweenMay 2016 and
June 2017. Practically, all pregnant women in Finland visit
a public health nurse regularly in a maternal health clinic.
They may also participate in a free of charge ultrasound
examination at the end of first trimester. The participants of
this studywere recruited in this examination satisfying certain
criteria:
1) The participant is at least 18 years old.
2) She should expect her first child.
3) The pregnancy is singleton.
4) The gestational age should be less than 15 weeks.
5) She understands Finnish or English.
6) She owns a smartphone, tablet, or personal computer.
Twenty-two pregnant women who met the criteria were
informed after the ultrasound examination. Based on this
initial interest, the procedure and purpose of the study were
provided for the women with phone calls. Twenty women
agreed to participate in the study. In face-to-face meetings,
the researchers collected background information of the par-
ticipants, some of which presented in Table 1. Afterward,
the wearable devices and instructions were delivered to the
participants.
C. ETHICS
The study was conducted in accordance with the code of
ethics of the World Medical Association (Declaration of
Helsinki) for involving human subjects in the experiments.
It was also approved by the joint ethics committee of the
hospital district of Southwest Finland (35/1801/2016) and
Turku University Hospital (TYKS). Moreover, the written
informed consent was obtained from all participants enrolled.
In addition, the permission to use Garmin Vivosmart R© HR
(Garmin Ltd, Schaffhausen, Switzerland) in this study was
acquired from the manufacturer Garmin Ltd.
VOLUME 7, 2019 93437
I. Azimi et al.: Personalized Maternal Sleep Quality Assessment: An Objective IoT-based Longitudinal Study
IV. SLEEP QUALITY ANALYSIS
In this section, we present our sleep quality analysis approach
tailored for assessment of maternal sleep adaptations during
pregnancy and postpartum. From the collected data, we first
extract several sleep attributes, each of which focuses on a
specific aspect of sleep quality. Changes and trends of these
attributes are explored for each subject throughout the mon-
itoring process. We then propose a personalized sleep model
for each subject to assess sleep quality in a comprehensive
and personalized way. The personalized model is constructed
by feeding the sleep attributes from the early stages of the
monitoring to a machine learning approach.
A. SLEEP ATTRIBUTES
Various objective sleep attributes have been proposed in the
literature for sleep quality assessment at many levels [71].
The selection of these attributes depends on the type of
collected data (i.e., bio-signals and acceleration data) and
subsequently the level of the analysis. For example, actig-
raphy can be used to extract sleep quantity parameters such
as sleep duration and awake after sleep onset [72], [73].
On the other hand, EEG, EOG, and respiration signals are
utilized to obtain attributes related to the sleep stages (e.g.,
REM sleep) [74]. In this study, a wristband equipped with
PPG and IMU sensors is employed to continuously col-
lect different parameters such as physical activity, body
movements, and heart rates. We exploit these parameters
to extract conventional sleep quantity, quality, and schedule
attributes [17], [23], [71], [75]. In this regard, eight objective
sleep attributes are extracted from each sleep event dur-
ing nighttime to investigate maternal sleep adaptations. The
attributes are outlined as follows:
• Sleep Duration, also known as Total Sleep Time (TST),
indicates the total time that a user sleeps in a day [76].
It is one of the prevalent parameters in sleep analysis,
widely used as a predictor of illnesses and mortality.
The association between short/long sleep duration and
high risks of different diseases such as cardiovascular
diseases, stroke, and hypertension is demonstrated in
the literature [77], [78]. In this study, the sleep dura-
tion is extracted using sleep information (i.e., start and
end of the sleep) provided by the Garmin Vivosmart
HR. To validate the sleep information, we implemented
a manual cross-check between the sleep information
and other data such as body movements and heart
rates. The sleep information is corrected or discarded
if there was no match between the data. Note that a
Listwise deletion method is used to eliminate sleep
events including missing values [79]. We also excluded
short naps in the analysis, due to the limitations of our
study.
• Sleep Onset Latency (SOL) refers to the amount of time
that a user spends in bed before her status changes
to the sleep state [80]. In this study, the sleep onset
latency is obtained using the step counts data, and body
movements and orientations. It is the time between the
occurrence of the last step before the sleep event and the
beginning of the sleep event.
• Wake After Sleep Onset (WASO) refers to the amount of
time that a user is awake after the sleep has begun and
before the final awakening [80]. In this study, we use
body movements and orientations data to determine the
WASO during the sleep event. Step counts data are also
used to detect if the user leaves the bed.
• Sleep Fragmentation indicates the number of awaken-
ings that occur after the sleep is initiated and before the
final awakening [81]. In this study, the sleep fragmen-
tation is also obtained using the body movements and
step counts data, by counting the times the user wakes
or leaves the bed during the sleep event.
• Sleep Efficiency is the ratio of the time that the user is
sleeping (i.e., sleep duration) to the total time spent in
bed [4]. In this study, the bedtime is determined using
the step counts data. It is considered as the time between
the occurrence of the last step before the sleep event and
the first step after the sleep event. The sleep efficiency
is calculated as sleep duration divided by bedtime.
• Sleep Depth reflects the ratio of deep sleep duration (i.e.,
motionless sleep) to the amount of time of total sleep
(i.e., sleep duration). Conventionally, the sleep stages
including non-REM (i.e., N1, N2, N3, and N4 stages)
and REM sleep are measured via Polysomnography
tests utilizing EEG, EMG, and EOG signals [82], [83].
However, due to limitations of the data collection in
this long-term monitoring, these sleep stages cannot
be distinguished. In this study, this attribute is defined
according to the body movements data, showing the
amount of motionless sleep in total sleep period, which
likely reflects deep sleep (i.e., N3 and N4 stages).
• Resting Heart Rate refers to the number of heart beats
per minute when the user is at complete rest. As a
cardiovascular risk factor, this attribute was investigated
in studies, tackling associations between elevated resting
heart rate and increased risk of cardiovascular diseases
and mortality [84], [85]. In this study, we define this
attribute for each sleep period by calculating the mini-
mum value of total sleep heart rates.
• Heart Rate Recovery is the time between the start of the
sleep and the time when the resting heart rate is reached.
This attribute can be considered as a readiness score of
the user. In this study, heart rate recovery is obtained
using sleep event and resting heart rate information.
B. PERSONALIZED SLEEP MODEL
We propose a personalized sleep model to investigate sleep
quality adaptations in pregnancy and postpartum. The model
is trained via the user’s sleep data at the beginning of themon-
itoring. Then, the model is used to evaluate the changes and
trends of data from the rest of the monitoring (i.e., test data).
The test data instances are affected by the new life conditions
of pregnancy; and as the model output, a score is desirable
that is indicative of the degree of the sleep abnormality.
93438 VOLUME 7, 2019
I. Azimi et al.: Personalized Maternal Sleep Quality Assessment: An Objective IoT-based Longitudinal Study
The personalized models for sleep can leverage anomaly
detection methods for identifying such abnormalities and
outliers in a dataset. We delve into state-of-the-art anomaly
detectionmethods and develop a suitablemethod formaternal
sleep quality assessment. As mentioned in Section II-B, there
is a broad range of methods for anomaly detection. However,
many of them are inappropriate for our study.
In this monitoring scheme, a data instance or sleep event
is multivariate (i.e., multiple attributes), and no contextual or
behavioral data is included. Therefore, we only focus onPoint
Anomalies approaches where a data instance can be selected
as anomalous with respect to the rest of the data instances,
but not the context information. Moreover, the proposed
technique should create a model using the ‘‘normal’’ data.
Therefore, our selection is narrowed down to semi-supervised
anomaly detection techniques.
Considering the output produced by the anomaly detection,
binary techniques are not applicable in this study because
they assign a binary label (i.e., normal or abnormal) to the
test instance. Support vector machine-based methods are
examples of binary techniques. Also, rule-based techniques
generally require training data to contain labels for both nor-
mal and anomalous classes [55]. Moreover, Nearest Neigh-
bor techniques (e.g., KNN) use a distance between a test
data instance and its nearest neighbors to determine if it
is anomalous. However, their performance highly depends
on the size of the training data and dimensionality of the
features. Clustering techniques are difficult to apply when
the training data is small because there is a high tendency
for the anomalous class to form a large cluster leading to
a high false positive rate [61]. Statistical techniques present
alternatives that rely on the assumptions (i.e., statistical mod-
els) made about the data generating distribution. They are also
inappropriate since the assumptions tend not to hold true in
high-dimensional data (like our dataset) and cannot capture
interactions between features [51].
In contrast, artificial neural networks have been success-
fully applied to anomaly detection in various fields [53],
[58], [86]. Replicator Neural Networks (RNN), also known
as Auto-encoders, are the most commonly used form
of neural networks in semi-supervised and unsupervised
settings [58], [86], [87]. They are known for their ability to
work well with high dimensional datasets and to capture
linear and nonlinear interactions in the data. However, these
techniques might show poor performance when the training
data size is small.
Bayesian networks-based methods tackle this issue,
including probability distributions in their models. They pro-
vide an uncertainty estimate along with the output, where it
serves as a confidence bound on the output of the model.
In addition, the model performs efficiently in case of small
data instances and is robust to over-fitting [88]. This quality
is important in this study, as we have a limited amount
of data samples (i.e., sleep events for each participant) to
train an individualized sleep model. Integrating a Bayesian
method into artificial neural networks was first proposed by
FIGURE 2. Replicator neural network with one hidden layer.
MacKay [89] and Neal [90]. This technique has been applied
in several domains includingmedical diagnostics and Internet
traffic classification [91].
We exploit the same concept to construct the personalized
sleep model, incorporating a Bayesian approach into a Repli-
cator Neural Networks (RNN).
RNN was first proposed by Hawkins et al. [59] and has
been further developed by Dau et al. [60]. The method
belongs to the class of auto-associative Neural Networks
with compressed internal representations [60]. It captures a
nonlinear representation of the input data and attempts to
reproduce the input data as the output of the network. During
the training process, the weights in the network are optimized
to minimize reconstruction errors of the training data. For a
given data instance (i), the reconstruction error is defined as:
δi = 1n
n∑
j=1
(xij − oij)2 (1)
where n is the number of features in the data instance, xij is
the input data instance, and oij is the output of the RNN. The
reconstruction error, δi, can be used as the anomaly score for
the given data instance.
Our Bayesian RNN is designed with one hidden layer,
as depicted in Figure 2. Given the training inputs as
X = {x1, . . . , xn} and their corresponding outputs as Y =
{y1, . . . yn}, we aim to find a function, f w(X ) parameterized
by weights w, that is likely to generate the outputs. f w(x) is
defined as f w(X ) = g(W2h(X )), where h(X ) is the hidden
layer which is h(X ) = g(W1X ). W1 and W2 are weights
vectors defined over probability distributions; and the activa-
tion function is the rectified linear unit (ReLU) (i.e., g(z) =
max{0, z}).
It should be noted that Bayesian Neural Networks are
based on Bayes theorem, and in general we need to find the
posterior distribution of the weights. Therefore, we begin by
setting a prior probability distribution on the weights, p(w),
with a Gaussian probability distribution. We, then, obtain the
likelihood, p(Y |X ,w), by updating our beliefs about the prior,
p(w), after seeing the data and deciding which weights are
more likely to produce the outputs. The posterior distribution
p(w|X ,Y ) is defined over the space of the weights:
p(w|X ,Y ) = p(Y |X ,w)p(w)
p(Y |X ) (2)
VOLUME 7, 2019 93439
I. Azimi et al.: Personalized Maternal Sleep Quality Assessment: An Objective IoT-based Longitudinal Study
where p(Y |X ) is the model evidence. However, the posterior
distribution cannot be computed by Equation 2, as the model
evidence is intractable for most real life problems [88], [92].
Therefore, an approximation method such as Variational
Inference [93] is used to obtain an approximating distribution
as:
q(w) = p(Y |X ,w)p(w) (3)
q(w) should be as close as possible to the true posterior dis-
tribution p(w|X ,Y ) in Equation 2. Therefore, the Kullback–
Leibler (KL) divergence3 [94] of the two distributions must
be minimized:
KL(q(w)||p(w|X ,Y )) =
∫
q(w)log
(
q(w)
p(w|X ,Y )
)
dw (4)
However, Equation 4 still contains the model evidence, so it
is still intractable. This leads to the use of Evidence Lower
Bound (ELBO) as an alternative to the KL divergence. The
ELBO is the negative of the KL divergence up to a logarithm
constant. Therefore, maximizing the ELBO is equivalent to
minimizing theKL divergencewhich in turn lets us to approx-
imate the true posterior distribution:
ELBO =
∫
q(w)log p(Y |X ,w)dw− KL(q(w)||p(w))
≤ log p(Y |X ) (5)
In our Bayesian RNN, we maximize the objective in Equa-
tion 5. More details can be found in [88], [92], [95].
V. EXPERIMENTAL DETAILS AND RESULTS
Twenty pregnant women were recruited to participate in
this study. The gestational ages of the subjects were 12 ±
2.1 weeks at the beginning of the monitoring. On average,
the subjects were 25.7 years old and had pre-pregnancy body
mass index (BMI) of 25, with different lifestyles and back-
ground characteristics as shown in Table 1.
We excluded 7 participants from our sleep analysis, as they
forgot/refused to use the wristband during sleep, with the
result that their data were insufficient for our study. There-
fore, in the final analysis, 13 pregnant women were included
in our analysis. For these 13 subjects, we extracted valid sleep
data for 172.15 ± 33.29 days per person out of the total
216.61 ± 14.34 days of the monitoring (79.5%). The valid
sleep data included 76.08±15.17 days of the second trimester
per person, 78.69 ± 12.75 days of the third trimester, and
17.38± 10.45 days of 1-month postpartum.
Regular phone-interviews (i.e., once or twice a month)
were performed during the study to acquire subjective mea-
surements of their status. According to the self-reports,
the subjects mostly had their daily routines (i.e., regular
work or study) prior to week 30, and began maternity leaves
3KL divergence, written as KL(p||q) = ∫ p(x)log(log p(x)q(x))dx, is a
measure of the distance between probability distributions in this case p and
q. A known property of the KL divergence is that is always greater or equal
to zero
from weeks 30-34 through the end of our study. In addition,
the participants were requested to report if they encounter
sleep disturbances. On average, three women reported sleep
problems at each interview till week-34, and six women expe-
rienced difficulty at sleeping in the final weeks of pregnancy.
The complaints were mostly due to back pain, sickness, and
visiting the toilet during nights.
In the following, we first present the eight objective sleep
attributes measured from the participants during pregnancy
and the postpartum; then, we demonstrate the abnormality
scores calculated using our proposed approach.
A. SLEEP ATTRIBUTES
As discussed in Section IV-A, eight objective sleep attributes
are exploited in this study to investigate the maternal sleep
changes from different perspectives. To visualize the col-
lected data, we calculate the weekly average of the sleep
attributes, where each week contains valid sleep data for
at least 4 days. The weeks with less than 4-days data
were excluded (4.7 ± 3.6 weeks per person) to reduce the
bias.
The variations in attributes for the 13 participants are
illustrated in Figures 3, starting from week 13 to week 40 of
pregnancy and week 1 to week 4 of postpartum. The varia-
tions are depicted by minimum, first-quartile, median, third-
quartile, and maximum values of the attributes in each week.
Weeks 39, 40, and 41 were the delivery weeks of 3, 7, and
3 participants, respectively. We excluded the data of week
41 in the figures, since we had the sleep data of only one
participant.
Sleep duration, a key parameter in sleep quality assess-
ment, gradually decreased during pregnancy. As indicated
in Figure 3a, it was 8 hours and 20 minutes (median
value) on the weeks 13-15, then decreased by approxi-
mately 10% and 20% in the mid and end of third trimester,
respectively. It dropped to 5 hours and 50 minutes (median
value) on the first week of postpartum and increased
afterward.
On the other hand, the WASO dramatically increased (see
Figure 3b). This parameter was more than 2-times higher at
the third trimester and 3-times higher at the postpartum in
comparison to the second trimester. Therefore, the quality of
sleep diminished at the last stages of pregnancy, and it even
became worse after the labor.
Similarly, sleep fragmentation increased, so there were
more awakening times at the third trimester and postpartum as
illustrated in Figure 3c. The variations of the sleep efficiency
were in accordance with the previous attributes, where it
gradually decreased throughout the pregnancy and was at the
lowest after the delivery (see Figure 3d).
The increase in sleep onset latencywas insignificant during
pregnancy. As indicated in Figure 3e, the parameter slightly
elevated at the third trimester (on average 30.92 minutes) in
comparison to the second trimester (on average 27.69 min-
utes). In a similar manner, sleep depth hardly increased in the
pregnancy (see Figure 3f). However, the parameter jumped
93440 VOLUME 7, 2019
I. Azimi et al.: Personalized Maternal Sleep Quality Assessment: An Objective IoT-based Longitudinal Study
FIGURE 3. The sleep attributes of the 13 participants from week 13 to week 40 of pregnancy and week 1 to week 4 of postpartum. The variations
are indicated by minimum, first-quartile, median, third-quartile, and maximum values of the attributes.
to more than 40% after the labor. Accordingly, motionless
sleep (i.e., deep sleep) was relatively elevated in postpartum,
although the sleep duration was less than sleep duration in the
pregnancy period.
The heart-rate-related attributes are depicted in Figures 3g
and 3h.Resting heart rate increased in the second trimester by
more than 10%. However, the parameter was relatively less in
postpartum, where it was, on average, 55 beats per minute at
the postpartum week 4. As indicated in Figure 3h, heart rate
recovery also changed during pregnancy. It decreased in the
third trimester (on average 175.78 minutes) in comparison to
the second trimester (on average 201.71 minutes).
B. ABNORMALITY SCORE
Recall that the sleep quality score is computed through an
abnormality score using our Bayesian RNN approach. The
cloud server is responsible for the sleep model construction
(i.e., training phase) and abnormality score calculation (i.e.,
testing phase). To implement the Bayesian RNN, we use the
Lasagne [96] and PyMC3 [97] frameworks in Python. The
input data of the method are the sleep data. Each data instance
includes the eight sleep attributes of a sleep event during
nighttime. The method has one input, one output, and one
hidden layers, each of which has eight units (i.e., number of
the sleep attributes).
VOLUME 7, 2019 93441
I. Azimi et al.: Personalized Maternal Sleep Quality Assessment: An Objective IoT-based Longitudinal Study
1) MODEL CONSTRUCTION
As aforementioned, the training data are the ‘‘normal’’ data
in such semi-supervised algorithms. In this study, the user’s
sleep data at the beginning of the monitoring were considered
as the training data. These are the data from week 13 to
week 21, as the most similar data to the user’s normal
condition. It should be noted that, in an ideal situation,
pre-pregnancy sleep data should be selected as the training
dataset (i.e., ‘‘normal’’ data).
The training data were normalized and fed to the model.
Using the PyMC3, the weights were first initialized as normal
probability distributions and then were optimized by max-
imizing the Evidence Lower Bound from the Equation 5.
Therefore, the model was enabled to replicate the input train-
ing data at the output with the minimum error.
2) SCORE CALCULATION
The model, as a compressed representation of the training
dataset, was used to reconstruct the test data. In this study,
the test data were the sleep data from week 22 to the end of
the monitoring. The error of a test instance reconstruction
indicates the abnormality level of the test instance. Let us
take two different examples. 1) Themodel replicates the input
test data at the output with small error. This indicates the test
instance is close to the training dataset (i.e., a similar sample
was already seen in the training phase). Consequently, the test
instance is ‘‘normal’’. 2) The model reproduces the input test
data at the output with large error. This shows the test instance
is far from the training dataset (i.e., the instance is new to the
model). Therefore, it is ‘‘abnormal’’.
In this regard, the abnormality level (i.e., abnormality
score) is the distance between the input and reconstructed
output, calculated as:
s = 1
n
n∑
j=1
(xj − oj)2 (6)
where n is the number of sleep attributes which is 8, xj is
the original data instance and oj is the reconstructed data
instance.
In this work, a personalized RNN model was created
for each participant using her own data; and her test data
were evaluated with the personalizedmodel. The abnormality
scores of the 13 participants are shown in Figure 4, starting
fromweek 22. The overall median values gradually increased
as the pregnancy progresses. The highest scores during preg-
nancy were for week 35 to the labor. At the postpartum
week 1, the score jumped to more than 230% in comparison
to week 40. This means that the worst sleep quality was for
the first week after the labor. Afterward, the scores slightly
decreased in the postpartum although they were considerably
higher than the scores during the pregnancy.
VI. DISCUSSION AND EVALUATION
To the best of our knowledge, this is the first IoT-based longi-
tudinal study that objectively assesses maternal sleep quality
FIGURE 4. The abnormality scores of the 13 participants of pregnancy
weeks 22-40 and postpartum weeks 1-4.
during pregnancy and postpartum. This IoT-basedmonitoring
provides a feasible method to assess the quality of women’s
sleep in a challenging transition period from pregnancy to
motherhood. In this section, we first discuss the observations
made by analyzing each attribute individually and then look
into the final sleep abnormality score.
A. SLEEP ATTRIBUTES
Different objective sleep attributes indicate the quality of
sleep diminished during pregnancy and in postpartum. Com-
pared with the existing studies, this work represents a higher
confidence level on these findings by performing long-term
and fine-grained quantitative measurements and analysis of
everyday data of pregnant women.
We found that the sleep duration and sleep efficiency
gradually decreased across pregnancy. Correspondingly,
theWASO and sleep fragmentation increased. These findings
of this continuous wristband monitoring are in concordance
with previous knowledge gained from short-term measure-
ments in a few separate time points. Sleep disturbances
during pregnancy could be considered unavoidable due to
the hormonal, anatomical, and physiological changes in the
woman’s body. For example, the levels of oxytocin, prolactin,
and cortisol increase and have effects on sleep regulation.
Furthermore, respiratory, musculoskeletal, and cardiovascu-
lar changes, as well as weight gain and bladder compression
by the uterus have impacts on sleep [80].
Moreover, our results indicate there are more changes in
these attributes after the delivery. The sleep duration and sleep
efficiency drop by 21.5% and 9.7%, and theWASO and sleep
fragmentation increase by 3.5 and 4.7 times, in comparison to
the second trimester. These postpartum findings also comply
with the previous findings; the changed life situation is a com-
mon reason for such poor sleep quality. In a previous study
93442 VOLUME 7, 2019
I. Azimi et al.: Personalized Maternal Sleep Quality Assessment: An Objective IoT-based Longitudinal Study
by Hughes et al. [98], for example, the total sleep time in the
first 48 hours after birth was less than 10 hours; however,
breastfeeding mothers slept longer than bottle-feeding moth-
ers. Sleep is often compromised in the postpartum period
during the first months because of infants’ sleep-wake pat-
terns and various needs leading multiple night-time awaken-
ings. Total sleep time appears to be the lowest one month
after birth, but it can remain as low still at two months
postpartum [39], [99]. In previous studies, these attributes
were measured via subjective self-report questionnaires or
short-term objective actigraphy [5], [16], [31], [100].
Based on the data in this study, the sleep onset latency did
not change significantly during pregnancy; however, the dif-
ficulties of falling asleep have been reported to increase
as pregnancy progresses [101]. In [101], about one-fourth
of pregnant women have suffered from daytime sleepiness
which might be an indicator of the insufficient sleep depth.
Subjectively rated sleepiness symptoms remained the same
during pregnancy [101] as did the sleep depth in this study.
Interestingly, the sleep depth increased more than 40% after
the delivery. This might be explained with the sleep depth
accumulated during pregnancy. Findings related to the heart
rate were supported by the earlier knowledge [102]; resting
heart rate increased during pregnancy but decreased again
during the first month postpartum, and heart rate recovery
decreased toward the end of pregnancy.
B. ABNORMALITY SCORE
Each sleep attribute represents the maternal sleep quality
from a single perspective. We tackled this issue by using an
abnormality score which is the fusion of the sleep attributes.
It provides a better understanding of changes in maternal
individual sleep quality, tailoring sleep data of early preg-
nancy to evaluate sleep data of late pregnancy and post-
partum. In an ideal situation, changes would be evaluated
against pre-conception sleep quality [103]. Moreover, it can
be used to achieve personalized healthcare. The proposed
score enables personalized decision-making through objec-
tive sleep quality assessment, where the intensity of the score
corresponds to its distance from the user’s normal condition
(i.e., user’s model). This personalization is important in such
health-related applications, as the normal health condition is
specific for each individual and is not easy to be generally
defined. For example, average resting heart rates of two
different persons could be 50 and 60 beats/min, both of which
are normal values according to their individual conditions.
We evaluate the obtained abnormality scores, comparing
the proposed sleep model with a baseline method. Recall that
as a semi-supervised approach is used in this work, the train-
ing data are label as ‘‘normal’’ and the test data are unlabeled.
To evaluate the model, we rely on the general hypothesis
behind the model, which should produce a higher score in
the case of anomalous data (i.e., differentiate ‘‘normal’’ and
‘‘abnormal’’ test instances).
In this regard, we consider a simple aggregate method
as a baseline for the performance comparison. The baseline
FIGURE 5. The abnormality scores of two participants, using the baseline
and proposed methods.
method determines sleep quality scores using overall popu-
lation values in normal conditions. We use the data from the
beginning of the monitoring (i.e., normal data) representing
the most probable sleep attributes of normal conditions in our
study. Eventually, the baseline score of each sleep event is the
sum of distances between the sleep attributes and their corre-
sponding normal population means in units of the standard
deviations.
We select two participants (i.e., P1 and P2) with different
conditions to implement the comparison between the pro-
posed method and the baseline. P1 experienced substantial
changes in her sleep although P2 had relatively less sleep
changes in pregnancy. Table 2 shows average values of some
sleep attributes of P1 and P2 in their normal conditions (i.e.,
beginning of the monitoring) and at the end of the pregnancy.
The table also indicates attributes changes (ratio), comparing
data at the end of pregnancy to population data and to her own
data. As indicated, the ratio of P1 attributes to her own data
is higher than the ratio to the population data. On the other
hand, the ratio of P2 attributes to her own data is relatively
less.
As shown in Figure 5a, the baseline score is unable to accu-
rately distinguish between P1 and P2. This is because P1’s
sleep parameters, despite the substantial changes, were close
to the population values. In contrast to the baseline method,
the sleep changes are clearly visible using the abnormality
score obtained from the proposed model (see Figure 5b). This
enables the provision of tailored individualized and effective
care, where we can identify those who need the care most and
optimize resource allocation.
C. LIMITATIONS AND FUTURE DIRECTIONS
The proposed IoT-based system is a proof-of-concept for
1) long-term monitoring of maternal daily sleep 2) effec-
tive care for maternal sleep disturbances using personalized
VOLUME 7, 2019 93443
I. Azimi et al.: Personalized Maternal Sleep Quality Assessment: An Objective IoT-based Longitudinal Study
TABLE 2. P1 and P2 attributes and the ratio of the attributes at the end of
pregnancy to her own data and population values.
decision-making. One of the limitations of this study is that
the study sample is small. Other studies investigate the asso-
ciations between subjective sleep measurements and other
pregnancy-related parameters and complications on large
study samples. For example, Okun et al. [104] conduct a
study on 166 pregnant women via self-report questionnaires
and indicate that poor sleep quality is correlated with an
increased risk of preterm birth. Another study is performed
on 457 pregnant women to tackle the association between
sleep quality and type of delivery and length of the labor [22].
Unfortunately, we are unable to statistically investigate such
associations in our data since our sample size is smaller.
Future directions of this study are to perform objective lon-
gitudinal studies on a larger population focusing on such
correlations.
Another limitation of our monitoring study is linked to
the data collection. We were bounded to one wristband that
monitored heart rate, step counts, and body movements.
Future work will consider multimodal and multisensor data
collection and integration with more advanced sensor nodes,
enabling the capture of additional health/sleep attributes. For
instance, PPG as a non-invasive and convenient technique
can play a significant role in such monitoring systems [69].
Finger-based and wrist-based PPG sensors can be lever-
aged in this regard to continuously acquire different health
parameters such as heart rate variability and respiration rate.
Moreover, strap monitors can be employed to record EMG
signals for possible abdominal contractions extraction. How-
ever, to enhance the feasibility of long-termmonitoring, there
needs to be a balance between the number of wearables and
their continuous use, as a high number of wearable devices
could be impractical or inapplicable for sustained long-term
monitoring. For instance, in our study, despite using only one
wristband for the data collection, we were required to exclude
the sleep data of 7 participants out of 20 due to the high
volume of missing data. The main reasons were forgetfulness
and refusal of wearing the wristband during sleep.
Finally, it is worth noting that the proposed model can be
extended to contextual anomalies methods, considering the
contextual information. These longitudinal studies demand
remote and in-home monitoring in which the participants
might be involved in different conditions and environments.
Therefore, context information including personal lifelog-
ging data, ambient data, and medication reports can improve
the accuracy of the personalized decision-making.
VII. CONCLUSION
Maternal sleep quality alters during the pregnancy and post-
partum due to the adaptations of the maternal body. Such
variations in sleep should be closely monitored as poor sleep
quality might lead to various pregnancy complications. Con-
ventional studies are insufficient for this issue as they are
limited to restricted data collection approaches. In this paper,
we conducted an objective longitudinal study to thoroughly
investigate maternal sleep adaptations in pregnancy and post-
partum. We introduced an IoT-based system to remotely
monitor pregnant women 24/7. Several sleep attributes were
extracted to observe changes in maternal sleep patterns.
Moreover, we proposed a Bayesian RNN approach to con-
struct a personalized sleep model for each individual using
her own data. The sleep model was utilized to deliver an
abnormality score, which indicated the degree of maternal
sleep quality adaptations. In total, we collected 7 months of
data from 20 pregnant women; however, we only included
172.15 ± 33.29 days of valid sleep data per person from
13 pregnant women in our sleep analysis. For each sub-
ject, the sleep model was created using the data from the
beginning of the monitoring, and the model was tested on
the rest of the pregnancy and postpartum data. The obtained
scores showed that sleep abnormalities increased during the
pregnancy (2.87 times) and after the delivery (5.62 times)
in comparison to the mid of the second trimester. This work
indicated sleep quality decreased in pregnancy and postpar-
tum with a high confidence level, leveraging fine-grained
quantitative measurements and analysis on everyday data of
pregnant women.
ACKNOWLEDGMENT
The authors would like to thank Mr. Arman Anzanpour for
designing Figure 1.
REFERENCES
[1] P. Soma-Pillay, C. Nelson-Piercy, T. Heli and M. Alexandre, ‘‘Physio-
logical changes in pregnancy,’’ Cardiovascular J. Afr., vol. 27, no. 2,
pp. 89–94, 2016.
[2] K. Y. Lain and P. M. Catalano, ‘‘Metabolic changes in pregnancy,’’ Clin.
Obstetrics Gynecol., vol. 50, no. 4, pp. 938–948, Dec. 2007.
[3] G. Pien and R. Schwab, ‘‘Sleep disorders during pregnancy,’’ Sleep,
vol. 27, no. 7, pp. 1405–1417, Oct. 2004.
[4] K. Lee, M. E. Zaffke, and G. McEnany, ‘‘Parity and sleep patterns during
and after pregnancy,’’ Obstetrics Gynecol., vol. 95, no. 1, pp. 14–18,
Jan. 2000.
[5] C. Hedman, T. Pohjasvaara, U. Tolonen, A. S. Suhonen-Malm, and
W.Myllylä, ‘‘Effects of pregnancy onmothers’ sleep,’’ SleepMed., vol. 3,
no. 1, pp. 37–42, Jan. 2002.
[6] M. Manconi, V. Govoni, A. De Vito, N.T. Economou, E. Cesnik,
I. Casetta, G. Mollica, L. Ferini-Strambi, E. Granieri, ‘‘Restless legs
syndrome and pregnancy,’’ Neurology, vol. 63, no. 6, pp. 1065–1069,
2004.
[7] L. Hunter, J. D. Rychnovsky, and S. M. Yount, ‘‘A selective review of
maternal sleep characteristics in the postpartum period,’’ J. Obstetric,
Gynecologic Neonatal Nursing, vol. 38, no. 1, pp. 60–68, Jan. 2009.
[8] F. L. Facco, J. Kramer, K. H. Ho, P. C. Zee, and W. A. Grobman,
‘‘Sleep disturbances in pregnancy,’’ Obstetrics Gynecol., vol. 115, no. 1,
pp. 77–83, Jan. 2010.
93444 VOLUME 7, 2019
I. Azimi et al.: Personalized Maternal Sleep Quality Assessment: An Objective IoT-based Longitudinal Study
[9] M. L. Okun, J. M. Roberts, A. L. Marsland, and M. Hall, ‘‘How disturbed
sleep may be a risk factor for adverse pregnancy outcomes a hypothesis,’’
Obstetrical Gynecol. Surv., vol. 64, no. 4, p. 273, Apr. 2009.
[10] S. Reutrakul, N. Zaidi, K. Wroblewski, H. H. Kay, M. Ismail,
D. A. Ehrmann, and E. VanCauter, ‘‘Sleep disturbances and their relation-
ship to glucose tolerance in pregnancy,’’ Diabetes Care, vol. 34, no. 11,
pp. 2454–2457, Nov. 2011.
[11] L. Palagini, A. Gemignani, S. Banti, M. Manconi, M. Mauri, and
D. Riemann, ‘‘Chronic sleep loss during pregnancy as a determinant
of stress: Impact on pregnancy outcome,’’ Sleep Med., vol. 15, no. 8,
pp. 853–859, Aug. 2014.
[12] J. J. Chang, G. W. Pien, S. P. Duntley, and G. A. Macones, ‘‘Sleep
deprivation during pregnancy and maternal and fetal outcomes: Is there a
relationship?’’ Sleep Med Rev., vol. 14, no. 2, pp. 107–114, 2010.
[13] G. Bourjeily, C. A. Raker, M. Chalhoub, and M. A. Miller, ‘‘Pregnancy
and fetal outcomes of symptoms of sleep-disordered breathing,’’ Eur.
Respiratory J., vol. 36, no. 4, pp. 849–855, Oct. 2010.
[14] G. N. Pires, M. L. Andersen, M. Giovenardi, and S. Tufik, ‘‘Sleep
impairment during pregnancy: Possible implications on mother–infant
relationship,’’ Med. Hypotheses, vol. 75, no. 6, pp. 578–582, 2010.
[15] H. Skouteris, C. Germano, E. H. Wertheim, S. J. Paxton, and J. Milgrom,
‘‘Sleep quality and depression during pregnancy: A prospective study,’’
J. Sleep Res., vol. 17, no. 2, pp. 217–220, Jun. 2008.
[16] J. Mindell, R. A. Cook, and J. Nikolovski, ‘‘Sleep patterns and sleep
disturbances across pregnancy,’’ Sleep Med., vol. 16, no. 4, pp. 483–488,
Apr. 2015.
[17] G. Landry, J. R. Best, and T. Liu-Ambrose, ‘‘Measuring sleep quality
in older adults: A comparison using subjective and objective methods,’’
Front Aging Neurosci., vol. 7, p. 166, Sep. 2015.
[18] D. J. Buysse, C. F. Reynolds III, T. H. Monk, S. R. Berman, and
D. J. Kupfer, ‘‘The pittsburgh sleep quality index: A new instrument for
psychiatric practice and research,’’ Psychiatry Res., vol. 28, pp. 193–213,
May 1989.
[19] M. Thurtell, B. B. Bruce, D. B. Rye, N. J. Newman, and V. Biousse,
‘‘The Berlin questionnaire screens for obstructive sleep apnea in idio-
pathic intracranial hypertension,’’ J Neuroophthalmol, vol. 31, no. 4,
pp. 316–319, Dec. 2011.
[20] B. Gelaye, G. Addae, B. Neway, G. T. Larrabure-Torrealva, C. Qiu,
L. Stoner, M. A. Fernandez, S. E. Sanchez, and M. A. Williams, ‘‘Poor
sleep quality, antepartum depression and suicidal ideation among preg-
nant women,’’ J. Affect. Disorders, vol. 209, pp. 195–200, Feb. 2017.
[21] G. Li, L. Kong, H. Zhou, X. Kang, Y. Fang, and P. Li, ‘‘Relationship
between prenatal maternal stress and sleep quality in Chinese preg-
nant women: The mediation effect of resilience,’’ Sleep Med., vol. 25,
pp. 8–12, Sep. 2016.
[22] N. Zafarghandi, S. Hadavand, A. Davati, S. M. Mohseni,
F. Kimiaiimoghadam, and F. Torkestani, ‘‘The effects of sleep quality and
duration in late pregnancy on labor and fetal outcome,’’ J. Maternal-Fetal
Neonatal Med., vol. 25, no. 5, pp. 535–537, 2012.
[23] A. Harvey, K. Stinson, K. L. Whitaker, D. Moskovitz, and H. Virk,
‘‘The subjective meaning of sleep quality: A comparison of individ-
uals with and without insomnia,’’ Sleep, vol. 31, no. 3, pp. 383–393,
2008.
[24] J. Hartmann, C. E. Carney, A. Lachowski, and J. D. Edinger, ‘‘Exploring
the construct of subjective sleep quality in patients with insomnia,’’
J. Clin. Psychiatry, vol. 76, no. 6, pp. 768–773, 2015.
[25] L. Bakken, H. S. Kim, A. Finset, and A. Lerdal, ‘‘Subjective sleep quality
in relation to objective sleep estimates: Comparison, gender differences
and changes between the acute phase and the six-month follow-up after
stroke,’’ J. Adv. Nursing, vol. 70, no. 3, pp. 639–650, Mar. 2014.
[26] R. B. Berry and M. H. Wagner, ‘‘Polysomnography I,’’ in Sleep
Medicine Pearls, 3rd ed. Amsterdam, The Netherlands: Elsevier, 2014,
pp. 80–87.
[27] A. Axelin, J. Kirjavainen, S. Salanterä, and L. Lehtonen, ‘‘Effects of pain
management on sleep in preterm infants,’’ Eur. J. Pain, vol. 14, no. 7,
pp. 752–758, 2010.
[28] S. Ancoli-Israel, R. Cole, C. Alessi, M. Chambers, W. Moorcroft, and
C. P. Pollak, ‘‘The role of actigraphy in the study of sleep and circadian
rhythms,’’ Sleep, vol. 26, no. 3, pp. 342–392, 2003.
[29] K. Lee and C. Gay, ‘‘Sleep in late pregnancy predicts length of labor
and type of delivery,’’ Amer. J. Obstetrics Gynecol., vol. 191, no. 6,
pp. 2041–2046, 2004.
[30] S. Coo Calcagni, B. Bei, J. Milgrom, and J. Trinder, ‘‘The relationship
between sleep and mood in first-time and experienced mothers,’’ Behav.
Sleep Med., vol. 10, no. 3, pp. 167–179, Feb. 2012.
[31] S. Tsai, P.-L. Lee, J.-W. Lin, and C.-N. Lee, ‘‘Cross-sectional and lon-
gitudinal associations between sleep and health-related quality of life
in pregnant women: A prospective observational study,’’ Int. J. Nursing
Stud., vol. 56, pp. 45–53, 2016.
[32] A. Haney, D. J. Buysse, B. L. Rosario, Y.-F. Chen, and M. L. Okun,
‘‘Sleep disturbance and cardiometabolic risk factors in early pregnancy:
A preliminary study,’’ Sleep Med., vol. 15, no. 4, pp. 444–450, 2014.
[33] L. Atzori, A. Iera and G. Morabito, ‘‘The Internet of Things: A survey,’’
Comput. Netw., vol. 54, no. 15, pp. 2787–2805, Oct. 2010.
[34] J. Gubbi, R. Buyya, S. Marusic, and M. Palaniswami, ‘‘Internet of Things
(IoT): A vision, architectural elements, and future directions,’’ Future
Gener. Comput. Syst., vol. 29, no. 7, pp. 1645–1660, 2013.
[35] M. Chen, S. Mao, and Y. Liu, ‘‘Big data: A survey,’’Mobile Netw. Appl.,
vol. 19, no. 2, pp. 171–209, Apr. 2014.
[36] J. Catov and C. Margerison-Zilko, ‘‘Pregnancy as a window to future
health: Short term costs and consequences,’’Amer. J. Obstetrics Gynecol.,
vol. 215, no. 4, pp. 406–407, Oct. 2017.
[37] UNICEF. (2019). Antenatal Care—UNICEF DATA. [Online]. Available:
https://data.unicef.org/topic/maternal-health/antenatal-care/
[38] Y. Yang, J.Mao, Z. Ye, X. Zeng, H. Zhao, Y. Liu, and J. Li, ‘‘Determinants
of sleep quality among pregnant women in China: A cross-sectional sur-
vey,’’ J. Maternal-Fetal Neonatal Med., vol. 31, no. 22, pp. 2980–2985,
2017.
[39] B. Bei, S. Coo, and J. Trinder, ‘‘Sleep and mood during pregnancy and
the postpartum period,’’ Sleep Med. Clinics, vol. 10, no. 1, pp. 25–33,
Mar. 2015.
[40] B. I. Balserak andK. Lee, ‘‘Sleep disturbances and sleep-related disorders
in pregnancy,’’ in Principles and Practice of Sleep Medicine. Amsterdam,
The Netherlands: Elsevier, 2010, pp. 1572–1586.
[41] V. Ibáñez, J. Silva and O. Cauli, ‘‘A survey on sleep assessment methods,’’
PeerJ, vol. 6, May 2018, Art. no. e4849.
[42] H. Zhou, W. Li, and Y. Ren, ‘‘Poor sleep quality of third trimester exac-
erbates the risk of experiencing postnatal depression,’’ Psychol Health
Med., vol. 17, pp. 1–10, Nov. 2018.
[43] E. Paavonen, O. Saarenpää-Heikkilä, P. Pölkki, A. Kylliäinen,
T. Porkka-Heiskanen, and T. Paunio, ‘‘Maternal and paternal sleep
during pregnancy in the Child-sleep birth cohort,’’ Sleep Med., vol. 29,
pp. 47–56, Jan. 2017.
[44] S. J. Herring, G. D. Foster, G. W. Pien, K. Massa, D. B. Nelson,
P. R. Gehrman and A. Davey, ‘‘Do pregnant women accurately report
sleep time?’’ Sleep Breathing, vol. 17, no. 4, pp. 1323–1327, 2013.
[45] B. Vaughn and P. Giallanza, ‘‘Technical review of polysomnography,’’
Chest, vol. 134, no. 6, pp. 1310–1319, Dec. 2008.
[46] S. J. Herring, D. B. Nelson, G. W. Pien, C. Homko, L. M. Goetzl,
A. Davey and G. D. Foster, ‘‘Objectively measured sleep duration and
hyperglycemia in pregnancy,’’ Sleep Med., vol. 15, no. 1, pp. 51–55,
Jan. 2014.
[47] F. Lin, Y. Zhuang, C. Song, A. Wang, Y. Li, C. Gu, C. Li and
W. Xu, ‘‘SleepSense: A noncontact and cost-effective sleep moni-
toring system,’’ IEEE Trans. Biomed. Circuits Syst., vol. 11, no. 1,
pp. 189–202, Feb. 2017.
[48] M. Tenhunen, E. Elomaa, H. Sistonen, E. Rauhala and S. L. Himanen,
‘‘Emfit movement sensor in evaluating nocturnal breathing,’’ Respiratory
Physiol. Neurobiol., vol. 187, no. 2, pp. 183–189, Jun. 2013.
[49] S. Saeb, T. R. Cybulski, S. M. Schueller, K. P. Kording and D. C. Mohr,
‘‘Scalable passive sleep monitoring using mobile phones: Opportunities
and obstacles,’’ J. Med. Internet Res., vol. 19, no. 4, p. e118, 2017.
[50] Z. Liu, L. Jiang, W. Chen, K.-I. Kitamura, and T. Nemoto, ‘‘Character-
ization of maternal body movement during sleep before and after partu-
rition,’’ in Proc. Annu. Int. Conf. IEEE Eng. Med. Biol. Soc., Sep. 2011,
pp. 1503–1506.
[51] V. Chandola, A. Banerjee and V. Kuma, ‘‘Anomaly detection: A survey,’’
ACM Comput. Surv., vol. 41, no. 3, pp. 15:1–15:58, Jul. 2009.
[52] M. Ahmed, A. N. Mahmood, and J. Hu, ‘‘A survey of network anomaly
detection techniques,’’ J. Netw. Comput. Appl., vol. 60, pp. 19–31,
Jan. 2016.
[53] S. J. Han and S. B. Cho, ‘‘Evolutionary neural networks for anomaly
detection based on the behavior of a program,’’ IEEE Trans. Syst., Man,
Cybern., B, Cybern., vol. 36, no. 3, pp. 559–570, Jun. 2005.
[54] W. Chen, S.-H. Hsu, and H.-P. Shen, ‘‘Application of SVM and ANN for
intrusion detection,’’ Comput. Oper. Res., vol. 32, no. 10, pp. 2617–2634,
Oct. 2005.
[55] W. Fan, M. Miller, S. Stolfo, W. Lee and P. Chan, ‘‘Using artificial
anomalies to detect unknown and known network intrusions,’’ in Proc.
IEEE Int. Conf. Data Mining, Dec. 2001, pp. 123–130.
VOLUME 7, 2019 93445
I. Azimi et al.: Personalized Maternal Sleep Quality Assessment: An Objective IoT-based Longitudinal Study
[56] T. Fawcett and F. J. Provost, ‘‘Activity monitoring: Noticing interesting
changes in behavior,’’ in Proc. 5th ACM SIGKDD Int. Conf. Knowl.
Discovery Data Mining, 1999, pp. 53–62.
[57] L. Manevitz and M. Yousef, ‘‘One-class SVMs for document classifica-
tion,’’ J. Mach. Learn. Res., vol. 2, pp. 139–154, Dec. 2002.
[58] R. Chalapathy, A. K. Menon, and S. Chawla, ‘‘Anomaly detection using
one-class neural networks,’’ 2018, arXiv:1802.06360. [Online]. Avail-
able: https://arxiv.org/abs/1802.06360
[59] S. Hawkins, H. He, G. Williams and R. Baxter, ‘‘Outlier detection using
replicator neural networks,’’ in Proc. Int. Conf. DataWarehousing Knowl.
Discovery, Sep. 2002, pp. 170–180.
[60] H. Dau, V. Ciesielski, and A. Song, ‘‘Anomaly detection using replicator
neural networks trained on examples of one class,’’ in Proc. Asia–íPacific
Conf. Simulated Evol. Learn., 2014, pp. 311–322.
[61] S. Basu, M. Bilenko, and R. J. Mooney, ‘‘A probabilistic framework for
semi-supervised clustering,’’ in Proc. ACM SIGKDD Int. Conf. Knowl.
Discovery Data Mining, 2004, pp. 59–68.
[62] J. Lin, E. Keogh, A. Fu and H. Van Herle, ‘‘Approximations to magic:
Finding unusual medical time series,’’ inProc. 18th IEEE Symp. Comput.-
Based Med. Syst., Jun. 2005, pp. 329–334.
[63] L. Cadmus-Bertram, ‘‘Using fitness trackers in clinical research: What
nurse practitioners need to know,’’ J. Nurse Practitioners, vol. 13, no. 1,
pp. 34–40, Jan. 2017.
[64] M. de Zambotti, L. Rosas, I. M. Colrain and F. C. Baker, ‘‘The sleep of
the ring: Comparison of the o¯ura sleep tracker against polysomnography,’’
Behavioral Sleep Med., vol. 17, no. 2, pp. 124–136, 2019.
[65] S. Kang, J. M. Kang, K.-P. Ko, S.-C. Park, S. Mariani, and J. Weng,
‘‘Validity of a commercial wearable sleep tracker in adult insomnia
disorder patients and good sleepers,’’ J. Psychosomatic Res., vol. 97,
pp. 38–44, Jun. 2017.
[66] M. de Zambotti, F. C. Baker, I. M. Colrain, ‘‘Validation of sleep-tracking
technology compared with polysomnography in adolescents,’’ Sleep,
vol. 38, no. 9, pp. 1461–1468, Sep. 2015.
[67] Garmin. (2019). Vivosmart HR | Activity Tracker. [Online]. Available:
https://buy.garmin.com/en-US/US/p/531166
[68] K. Grym, H. Niela-Vilén, E. Ekholm, L. Hamari, I. Azimi, A. Rahmani,
P. Liljeberg, E. Löyttyniemi and A. Axelin, ‘‘Feasibility of smart wrist-
bands for continuous monitoring during pregnancy and one month after
birth,’’ BMC Pregnancy Childbirth, vol. 19, no. 34, p. 34, Jan. 2019.
[69] J. Allen, ‘‘Photoplethysmography and its application in clinical physio-
logical measurement,’’ Physiol Meas., vol. 28, no. 3, pp. 1–39, Feb. 2007.
[70] Linode. (2019). Ssd Cloud Hosting& Linux Servers. [Online]. Available:
https://www.linode.com/
[71] A. Sadeh, ‘‘III. Sleep assessment methods,’’Monographs Soc. Res. Child
Develop., vol. 80, no. 1, pp. 33–48, Mar. 2015.
[72] L. Tikotzky and E. Volkovich, ‘‘Infant nocturnal wakefulness: A longitu-
dinal study comparing three sleep assessment methods,’’ Sleep, vol. 42,
no. 1, p. zsy191, 2018.
[73] V. Natale, G. Plazzi, and M. Martoni, ‘‘Actigraphy in the assessment of
insomnia: A quantitative approach,’’ Sleep, vol. 32, no. 6, pp. 767–771,
2009.
[74] I. Fonareva, A. M. Amen, D. P. Zajdel, R. M. Ellingson, and B. S. Oken,
‘‘Assessing sleep architecture in dementia caregivers at home using an
ambulatory polysomnographic system,’’ J. Geriatric Psychiatry Neurol.,
vol. 24, no. 1, pp. 50–59, 2011.
[75] A. D. Krystal and J. D. Edinger, ‘‘Measuring sleep quality,’’ Sleep Med.,
vol. 9, no. 1, p. s10–s17, Sep. 2008.
[76] C. Kline, ‘‘Sleep duration,’’ in Encyclopedia of Behavioral Medicine.
New York, NY, USA: Springer, 2013, pp. 1808–1810.
[77] M. Jike, O. Itani, N. Watanabe, D. J. Buysse, and Y. Kaneita, ‘‘Long sleep
duration and health outcomes: A systematic review, meta-analysis and
meta-regression,’’ Sleep Med. Rev., vol. 39, pp. 25–36, Jun. 2018.
[78] A. N. Vgontzas, D. Liao, E. O. Bixler, G. P. Chrousos and A. Vela-Bueno,
‘‘Insomnia with objective short sleep duration is associated with a high
risk for hypertension,’’ Sleep., vol. 32, no. 4, pp. 491–497, 2009.
[79] X. H. Zhou, C. Zhou, D. Lui and X. Ding, Applied Missing Data Analysis
in the Health Sciences. Hoboken, NJ, USA: Wiley, 2014.
[80] B. Balserak and K. Lee, ‘‘Sleep and sleep disorders associated with
pregnancy,’’ in Principles and Practice of Sleep Medicine. Amsterdam,
The Netherlands: Elsevier, 2017, pp. 1525–1539.
[81] E. Mezick, ‘‘Sleep fragmentation,’’ in Encyclopedia of Behavioral
Medicine. New York, NY, USA: Springer, 2013, pp. 1810–1811.
[82] C. Kline, ‘‘Sleep quality,’’ in Encyclopedia of Behavioral Medicine.
New York, NY, USA: Springer, 2013, pp. 1811–1813.
[83] S. Insana, ‘‘Non-REM sleep,’’ in Encyclopedia of Behavioral Medicine.
New York, NY, USA: Springer, 2013, pp. 1342–1343.
[84] M. T. Cooney, E. Vartiainen, T. Laakitainen, A. Juolevi, A. Dudina and
I. M. Graham, ‘‘Elevated resting heart rate is an independent risk factor
for cardiovascular disease in healthy men and women,’’ Amer. Heart J.,
vol. 159, no. 4, pp. 612–619, Apr. 2010.
[85] M. Böhm, J. C. Reil, P. Deedwania, J. B. Kim and J. S. Borer, ‘‘Resting
heart rate: Risk indicator and emerging risk factor in cardiovascular
disease,’’ Amer. J. Med., vol. 128, no. 3, pp. 219–228, Mar. 2015.
[86] C. Zhang, D. Song, Y. Chen, X. Feng, C. Lumezanu, W. Cheng,
J. Ni, B. Zong, H. Chen and N. V. Chawla, ‘‘A deep neural net-
work for unsupervised anomaly detection and diagnosis in multivari-
ate time series data,’’ 2018, arXiv:1811.08055. [Online]. Available:
https://arxiv.org/abs/1811.08055
[87] C. Zhou and R. Paffenroth, ‘‘Anomaly detection with robust deep autoen-
coders,’’ in Proc. ACM Int. Conf. Knowl. Discovery Data Mining,
Jul. 2017, pp. 665–674.
[88] Y. Gal, ‘‘Uncertainty in deep learning,’’ Ph.D. dissertation, Dept. Eng.,
Univ. of , Cambridge, U.K., 2016.
[89] D. J. MacKay, ‘‘A practical Bayesian framework for backpropagation
networks,’’ Neural Comput., vol. 4, no. 3, pp. 448–472, 1992.
[90] R. Neal, Bayesian Learning for Neural Networks. New York, NY, USA:
Springer, 2012.
[91] T. T. T. Nguyen and G. Armitage, ‘‘A survey of techniques for Internet
traffic classification using machine learning,’’ IEEE Commun. Surveys
Tuts., vol. 10, no. 4, pp. 56–76, Apr. 2008.
[92] D. Blei, A. Kucukelbir, and D. McAuliffe, ‘‘Variational inference:
A review for statisticians,’’ J. Amer. Stat. Assoc., vol. 112, no. 518,
pp. 859–877, 2017.
[93] M. I. Jordan, Z. Ghahramani, T. S. Jaakkola and L. K. Saul, ‘‘An introduc-
tion to variational methods for graphical models,’’Mach. Learn., vol. 37,
no. 2, pp. 183–233, 1999.
[94] S. Kullback and R. Leibler, ‘‘On information and sufficiency,’’Ann.Math.
Statist., vol. 22, no. 1, pp. 79–86, 1951.
[95] A. Kucukelbir, D. Tran, R. Ranganath, A. Gelman and D. M. Blei,
‘‘Automatic differentiation variational inference,’’ J. Mach. Learn. Res.,
vol. 18, no. 1, pp. 430–474, 2017.
[96] Lasagne Contributors. (2018). Lasagne 0.2.Dev1 Documentation.
[Online]. Available: https://lasagne.readthedocs.io/en/latest/
[97] PyMC Development Team. (2018). PyMC3 Documentation. [Online].
Available: https://docs.pymc.io/
[98] O. Hughes, M. M. Mohamad, P. Doyle and G. Burke, ‘‘The significance
of breastfeeding on sleep patterns during the first 48 hours postpartum
for first time mothers,’’ J. Obstet. Gynaecol., vol. 38, no. 3, pp. 316–320,
2018.
[99] J. J. Doering, A. Szabo, D. Goyal and E. Babler, ‘‘Sleep quality and
quantity in low-income postpartum women,’’ MCN. Amer. J. Maternal
Child Nursing, vol. 42, no. 3, pp. 166–172, 2017.
[100] R. M. Román-Gálvez, C. Amezcua-Prieto, I. Salcedo-Bellido,
J. M. Martínez-Galiano, K. S. Khan and A. Bueno-Cavanillas, ‘‘Factors
associated with insomnia in pregnancy: A prospective cohort study,’’ Eur.
J. Obstetrics Gynecol. Reproductive Biol., vol. 221, pp. 70–75, 2018.
[101] P. Polo-Kantola, L. Aukia, H. Karlsson, L. Karlsson and E. J. Paavonen,
‘‘Sleep quality during pregnancy: Association with depressive and anxi-
ety symptoms,’’ Acta Obstetricia Et Gynecologica Scandinavica, vol. 96,
no. 2, pp. 198–206, 2017.
[102] L. Wolfe and T. Weissgerber, ‘‘Clinical physiology of exercise in preg-
nancy: A literature review,’’ J. Obstetrics Gynaecol. Canada, vol. 25,
no. 6, pp. 473–483, 2003.
[103] I. D. Sedov, E. E. Cameron, S. Madigan and L. M. Tomfohr-Madsen,
‘‘Sleep quality during pregnancy: A meta-analysis,’’ Sleep Med. Rev.,
vol. 38, pp. 168–176, 2018.
[104] M. L. Okun, C. D. Schetter and L. M. Glynn, ‘‘Poor sleep quality is
associated with preterm birth,’’ Sleep, vol. 34, no. 11, pp. 1493–1498,
2011.
IMAN AZIMI received the B.Sc. degree in
biomedical engineering from the University of
Isfahan, Iran, in 2010, and theM.Sc. degree in arti-
ficial intelligence and robotics from the Sapienza
University of Rome, Italy, in 2014. He is cur-
rently pursuing the Ph.D. degree with the Depart-
ment of Future Technologies, University of Turku,
Finland. He has authored more than 20 peer-
reviewed publications, both in medical and tech-
nological venues. His current research interests
include personalized health data analytics, remote health monitoring, the
Internet of Things, and embedded systems.
93446 VOLUME 7, 2019
I. Azimi et al.: Personalized Maternal Sleep Quality Assessment: An Objective IoT-based Longitudinal Study
OLUGBENGA OTI received the B.Sc. degree in
computer science fromBowenUniversity, Nigeria,
and theM.Sc. degree in computer science from the
University of Copenhagen, Denmark. During the
writing of this paper, she was a Research Assis-
tant with the University of Turku, Finland. Her
research interests include machine learning, big
data analytics, and computational biology.
SINA LABBAF received the bachelor’s degree
in computer engineering from the University of
Tehran, Iran, in 2017. He is currently pursuing
the Ph.D. degree in computer science with the
Donald Bren School of Information and Computer
Science, University of California at Irvine. His
research interests include the Internet of Things
for health care, biological data analytics, and con-
nected health services.
HANNAKAISA NIELA-VILÉN is currently a
Postdoctoral Researcher with the Department of
Nursing Science, University of Turku, Finland.
She works as part of a research program called
Health in Early Life and Parenthood. Before aca-
demic career, she has worked seven years as a
midwife in the Labor and Delivery Unit, Turku
University Hospital. Her current research projects
are about the possibilities of remote monitoring in
maternity care, early contact between amother and
her newborn infant, and breastfeeding.
ANNA AXELIN received the Ph.D. degree from
the University of Turku, Finland, in 2010. Her aca-
demic career has included conducting quantitative
and qualitative research on maternity and neonatal
care inmultidisciplinary and international research
groups. She did her Postdoctoral training in the
Department of Family Health Care Nursing, Uni-
versity of California San Francisco. In 2018, she
was appointed as an Associate Researcher with the
Department of International Maternal and Child
Health, University of Uppsala, Sweden. In addition to the academic career,
she has ten-year working experience as an NICU Nurse. She is leading the
Health in Early Life and Parenthood (HELP) Research Group, which aims
to promote health and welfare in the early stages of life. She is currently
an Associate Professor with the Department of Nursing Science, University
of Turku, Finland. Her special research interests include how to keep parents
and sick newborns together throughout the infant hospital stay and strengthen
their relationship already during pregnancy, pain and sleep in neonates, and
the implementation of evidence-based practice inmaternity and neonatal care
with the help information technology.
NIKIL DUTT received the Ph.D. degree in com-
puter science from the University of Illinois at
Urbana–Champaign, in 1989. He is currently a
Distinguished Professor of computer science, cog-
nitive sciences, and EECS with the University of
California at Irvine. He is also a Distinguished
Visiting Professor with the CSE Department, IIT
Bombay, India. He has coauthored seven books on
topics covering hardware synthesis, memory and
computer architecture specification and validation,
and on-chip networks. His research interests include embedded systems,
electronic design automation (EDA), computer systems’ architecture and
software, the healthcare IoT, and brain-inspired architectures and computing.
He is a Fellow of the IEEE and the ACM.Hewas a recipient of the IFIP Silver
Core Award. He received over a dozen best paper awards and nominations
at premier EDA and embedded systems conferences. He has served as the
Editor-in-Chief of the ACM TODAES and as an Associate Editor for the
ACM TECS and the IEEE TVLSI. He has extensive service on the steering,
organizing, and program committees of several premier EDA and embedded
system design conferences and workshops, and also serves or has served on
the advisory boards of ACM SIGBED, ACM SIGDA, ACMTECS, the IEEE
EMBEDDED SYSTEMS LETTERS (ESL), and the ACM Publications Board.
PASI LILJEBERG received the M.Sc. and Ph.D.
degrees in electronics and information technol-
ogy from the University of Turku, Turku, Finland,
in 1999 and 2005, respectively. He received the
Adjunct Professorship in embedded computing
architectures, in 2010. He is currently a Professor
in embedded systems and Internet of Things with
the University of Turku. In that context, he has
established and leading the Internet-of-Things for
Healthcare (IoT4Health) Research Group. He has
authored around 300 peer-reviewed publications. His current research inter-
ests include biomedical engineering and health technology.
AMIR M. RAHMANI received the M.Sc. degree
from the University of Tehran, Iran, in 2009,
the Ph.D. degree from the University of Turku,
Finland, in 2012, and the M.B.A. degree jointly
from the Turku School of Economics and the Euro-
pean Institute of Innovation and Technology Digi-
tal, in 2014. He was a Marie Curie Global Fellow
with the University of California at Irvine, Irvine,
USA, and TU Wien, Austria, from 2016 to 2019.
He is currently an Assistant Professor with the
University of California at Irvine. He is also an Adjunct Professor (Docent)
with the University of Turku. His work spans mobile health, the wear-
able Internet-of-Things, self-aware computing, health informatics, and fog
computing. He has authored over 190 peer-reviewed publications. He is an
Associate Editor of the ACM Transactions on Computing for Healthcare.
VOLUME 7, 2019 93447