ArticlePDF Available

Automated Detection of Infant Holding Using Wearable Sensing: Implications for Developmental Science And Intervention

Authors:

Abstract and Figures

Physical contact is critical for children's physical and emotional growth and well-being. Previous studies of physical contact are limited to relatively short periods of direct observation and self-report methods. These methods limit researchers' understanding of the natural variation in physical contact across families, and its specific impacts on child development. In this study we develop a mobile sensing platform that can provide objective, unobtrusive, and continuous measurements of physical contact in naturalistic home interactions. Using commercially available motion detectors, our model reaches an accuracy of 0.870 (std: 0.059) for a second-by-second binary classification of holding. In addition, we detail five assessment scenarios applicable to the development of activity recognition models for social science research, where required accuracy may vary as a function of the intended use. Finally, we propose a grand vision for leveraging mobile sensors to access high-density markers of multiple determinants of early parent-child interactions, with implications for basic science and intervention.
Content may be subject to copyright.
64
Automated Detection of Infant Holding Using Wearable Sensing:
Implications for Developmental Science and Intervention
XUEWEN YAO, The University of Texas at Austin
THOMAS PLÖTZ, Georgia Institute of Technology
MCKENSEY JOHNSON, The University of Texas at Austin
KAYA DE BARBARO, The University of Texas at Austin
Physical contact is critical for children’s physical and emotional growth and well-being. Previous studies of physical contact
are limited to relatively short periods of direct observation and self-report methods. These methods limit researchers’
understanding of the natural variation in physical contact across families, and its specic impacts on child development. In this
study we develop a mobile sensing platform that can provide objective, unobtrusive, and continuous measurements of physical
contact in naturalistic home interactions. Using commercially available motion detectors, our model reaches an accuracy
of 0.870 (std: 0.059) for a second-by-second binary classication of holding. In addition, we detail ve assessment scenarios
applicable to the development of activity recognition models for social science research, where required accuracy may vary
as a function of the intended use. Finally, we propose a grand vision for leveraging mobile sensors to access high-density
markers of multiple determinants of early parent-child interactions, with implications for basic science and intervention.
CCS Concepts:
Human-centered computing Ubiquitous and mobile computing design and evaluation methods
;
Empirical studies in ubiquitous and mobile computing;
Additional Key Words and Phrases: Mother-infant Interaction, Attachment, Infant Holding, Wearable Sensor, Accelerometer,
Assessment Scenarios
ACM Reference Format:
Xuewen Yao, Thomas Plötz, McKensey Johnson, and Kaya de Barbaro. 2019. Automated Detection of Infant Holding Using
Wearable Sensing: Implications for Developmental Science and Intervention. Proc. ACM Interact. Mob. Wearable Ubiquitous
Technol. 3, 2, Article 64 (June 2019), 17 pages. https://doi.org/10.1145/3328935
1 INTRODUCTION
Research on the functions and importance of physical contact lags vastly behind that of other sensory modalities.
For example, there are nearly 13 times more publications on vision than touch [
22
]. While the importance of
physical contact is evident across the lifespan, it appears to play a particularly important role during infancy,
with benets spanning a number of domains.
Touch enhances infant physiological functioning, with meaningful implications for physical health. A meta-
analysis of 21 randomized control design studies involving over 3042 infants [
9
] indicated that increasing
skin-to-skin physical contact between parents and their preterm infants reduces rates of infant mortality (risk
ratio (RR): 0.67) and infection (RR: 0.5), and can shorten the duration of hospital stays (typical mean dierence:
Authors’ addresses: X. Yao, Department of Electrical and Computer Engineering, The University of Texas at Austin; T. Plötz, School of
Interactive Computing, Georgia Tech; M. Johnson and K. de Barbaro, Department of Psychology, The University of Texas at Austin.
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that
copies are not made or distributed for prot or commercial advantage and that copies bear this notice and the full citation on the rst
page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy
otherwise, or republish, to post on servers or to redistribute to lists, requires prior specic permission and/or a fee. Request permissions from
permissions@acm.org.
©2019 Association for Computing Machinery.
2474-9567/2019/6-ART64 $15.00
https://doi.org/10.1145/3328935
Proc. ACM Interact. Mob. Wearable Ubiquitous Technol., Vol. 3, No. 2, Article 64. Publication date: June 2019.
64:2 X. Yao et al.
0.9 day). Other forms of touch, such as infant massage, also appear to have similar functions. For example, a
review of nine studies of massage therapy indicated that preterm newborns who received 5-10 days of massage
therapy showed a 21-48% gain in weight and 3-6 days shorter hospital stay relative to preterm neonates who
received standard care [
17
]. Physical contact is thought to improve health outcomes via a variety of mechanisms.
For example, touch can promote the development of the parasympathetic nervous system, including enhancing
infant respiratory and thermoregulatory functioning [6].
Touch also appears to play an important role in establishing caregiver-infant bonding and attachment. In a
randomized control study, mother-infant pairs receiving two hours of skin-to-skin contact immediately following
their infants’ birth showed enhanced patterns of reciprocity, maternal sensitivity, and infant self-regulation one
year later relative to those separated for 2 hours after birth in a "standard care" condition [8].
While the benets of massage and skin-to-skin contact in the earliest weeks of life are relatively well-established,
clothed physical contact, experienced when a caregiver simply holds their child, also appears to have important
developmental consequences. In particular, both skin-to-skin and clothed holding behaviors are thought to play a
key role in the regulation of infants’ arousal and emotional state [
16
,
21
]. Parents’ participation in such cycles of
arousal and regulation during infants’ rst year is critical for the development of infants’ independent ability to
regulate their emotions, termed self-regulation, as well as the attachment relationship between children and their
caregivers. Both self-regulation and attachment relationships emerge within the rst year of life and together are
a foundation for social-emotional functioning across the lifespan, with implications for the development of mental
health disorders as well as numerous related behaviors across the lifespan[
4
,
41
]. For example, attachment style
classied at 12 months of age has been shown to predict high-quality play and exploration behaviors in the rst
year, better problem-solving, sociability, and independence in toddlers, and more curiosity, exible management
of behavior, and ego control in the preschool years [13,40].
Parent holding behaviors within the rst year contribute directly and indirectly to the regulation of infants’
distress. Physical touch is known to function as a pain analgesic for infants [
18
]. Indirectly, the physical act
of holding also allows caregivers to be more attentive to their infants’ needs. For example, while being held,
infants can indicate to their caregivers that they are hungry by rooting or clawing at them, thereby avoiding a
full-blown distress episode [
31
]. A small number of experimental studies have shown that holding can promote
infant regulation and increase attachment security. For example, a supplemental carrying intervention was found
to reduce infant crying in the rst 3 months of life by 43% [
24
]. In another intervention study [
2
], caregivers
of young infants were randomly assigned to either an experimental holding condition or a control condition.
Caregivers in the experimental condition received soft carriers which promoted physical contact while those
in the control condition received infant seats. Mothers in the experimental condition were more responsive to
their infants’ vocalizations at 3.5 months and their infants were more securely attached at 13 months, two key
indicators of adaptive social-emotional development.
While suggestive, research on the impacts of clothed physical contact is relatively limited, in large part due
to the lack of a consistent way to assess the duration and timing of holding behaviors as they occur in natural
settings. Indeed, research on the role of touch in infancy is typically conducted in hospital settings where
research participants can be directly monitored [
8
,
28
]. Alternatively, it relies on self-report diaries lled out by
participating caregivers[
24
]. This means that we have little understanding of the natural variations in the amount
of physical contact behaviors between families, and the broader impacts of those variations on child development.
More broadly, similar issues plague research on many aspects of early interaction between caregivers and their
children. While many aspects of daily activity are hypothesized to have consequences for children’s development,
most studies can only capture approximations of these behaviors in the laboratory.
The widespread adoption and presence of mobile and wearable sensors, paired with the coming of age
of powerful algorithms to automatically extract meaningful activities from raw sensor data, could provide
unprecedented access to the daily contexts in which development happens. Sensor platforms could be used
Proc. ACM Interact. Mob. Wearable Ubiquitous Technol., Vol. 3, No. 2, Article 64. Publication date: June 2019.
Automated Detection of Infant Holding Using Wearable Sensing: Implications for Developmental ... 64:3
to simultaneously capture the activity of the child and the caregiver, in addition to a vast array of possible
determinants of their actions: from their perceived environments and internal states, to ecological factors such as
household chaos [
11
]. This is critical as it is widely believed that developmental outcomes are not determined by
a single factor, such as genes or caregiving, but are rather the outcome of a complex dynamic ecosystem [
41
].
Captured repeatedly over months or years, high-density data on these various determinants of development
could provide a radical new opportunity to understand how and why dierences between individuals develop -
from school success to physical and mental health.
The ultimate goal of our research is to develop a sensors-to-analytics platform to capture high-density markers
of parent-child activity "in the wild" i.e. as they go about their typical day-to-day activities. Such a platform
will provide unprecedented access to objective, unobtrusive, high-density measurements of developmental
determinants, pushing the limits of basic developmental psychology research. In particular, such novel datasets
have the potential to lead to a better understanding of children’s individual trajectories of risk and resilience,
with the potential for earlier diagnosis of potential issues and abnormalities as well as opportunities for timely
intervention and improved care. Wearable and ubiquitous computing methods have the potential to play a key
role in this endeavor, serving as enabling technology that facilitates systematic and automated assessments of
natural behavior on a large scale.
In this paper, we lay the foundations for our research agenda. Specically, we pursue one aspect of the
envisioned framework, namely, the automated assessment of holding behavior - a highly critical yet understudied
component of early mother-infant interactions. The contributions of this paper are as follows:
We present the rst attempt, to our knowledge, to build a comprehensive system to automatically detect
multiple dimensions of caregiver-infant interactions. Our system will provide unique access to early
caregiving behaviors for developmental scientists while also advancing ubiquitous computing research via
a novel multi-person use case.
We developed a model to detect mother-infant holding behaviors using body-worn accelerometers. We
trained and evaluated our model with 26 mother-infant pairs, wearing our sensor for on average 45 minutes
(std =11) during naturalistic interactions. Our model successfully distinguishes holding from non-holding
activities with an accuracy of 0.870 (std: 0.059) at second-by-second resolution.
We present and discuss the accuracy of our model within ve specic assessment scenarios, including
event-based accuracy and comparisons of absolute and relative activity summaries across participants.
These additional assessment scenarios are of particular relevance to the developmental science community,
and more generally to social-science approaches interested in understanding individual dierences.
In a sub-sample of participants, we assessed the accuracy of our infant-holding detection model with sensors
placed on the wrist rather than chest. Our results indicate that chest-worn sensors are more accurate in all
assessment scenarios. However, wrist-worn sensors perform adequately for some assessments and could be
used for longer-term monitoring as they are more comfortable for participants. We discuss these usability
trade-os in the context of survey results detailing participant’s comfort in our discussion.
2 MOTIVATION AND BACKGROUND
Much of the research on the developmental impacts of touch and physical contact is limited to direct observations
of families during hospital stays (common when infants are born premature and must remain under supervision).
This means we know relatively little about the role of touch in healthy infants (i.e. those not born premature) as
well as older infants (who are no longer in the hospital). Another common approach within holding research
is to compare developmental outcomes of children of families randomly assigned to a "holding intervention"
vs. a "usual care" control intervention. Families in the holding intervention are provided a baby carrier in order
to encourage holding behaviors, whereas families in the control condition receive an infant seat [
2
]. While
Proc. ACM Interact. Mob. Wearable Ubiquitous Technol., Vol. 3, No. 2, Article 64. Publication date: June 2019.
64:4 X. Yao et al.
end-of-study survey measures have been used to validate that families assigned carriers report the use of carriers
more than the control families, this method provides no actual indication of the amount of time infants in either
condition are carried. This limits both the interpretability of such research as well as its accuracy, as actual time
spent holding is unknown.
Self reports are the gold-standard method for assessing holding in home environments. However, self-report
diaries are burdensome on families, meaning they are conducted only for short periods of time and they are
prone to errors and biases. Additionally, they provide highly imprecise data on the timing of physical contact.
This is critical as a key hypothesis of attachment theory is that soothing contact temporally contingent to infant
distress serves to regulate (i.e. reduce) infant arousal [
41
], providing key input for developing stress and arousal
systems [10] as well as infants’ expected patterns of distress and regulation.
Wearable sensors have the potential to greatly enhance the status quo for assessment of physical contact.
We hypothesize that it will be possible to leverage body-worn inertial sensors to automatically and objectively
capture parent-child holding behaviors with great precision. Activity recognition can be considered one of the
pillars of wearable and ubiquitous computing, and a large number of systems and methods have been developed
that successfully demonstrated the feasibility of activity recognition with wearable motion sensors specically in
health applications (e.g., [3,19,23,29,37,38]).
Despite the great popularity of wearable sensors and their wide usage in many applications, within the
ubiquitous computing community, research involving infants is not common. Exceptions are mainly focused on
medical applications. For example, Hayes et al. [
20
], developed motion sensing systems to support premature babies
in the transition from hospital to home. Additionally, Fan et al. [
14
] developed a Markov model based technique
for recognizing gestures of Cramped Synchronized General Movements which were highly correlated with an
eventual diagnosis of Cerebral Palsy. [
35
] described a child activity recognition approach using accelerometer
and barometer to prevent child accidents such as unintentional injuries.
Additionally, there is a consumer market for wearable devices that can detect infant activity and report it to
parents (as critically reviewed by [
43
]). These devices typically assess infant physical motion and vitals, including
blood oxygenation, temperature and breathing rate. However, these applications only begin to scratch the surface
of the potential and useful tools that could be developed. Thus, more research into reliable methods to objectively
access parent-child interactions is warranted.
2.1 Research Design and Sensor Selection
Our overall aim is to develop a platform to collect objective high-density markers of mother-infant interaction. In
particular, this platform should have the following functionalities. First, we wish to be able to detect markers
related to caregiving activities (such as physical contact, feeding, sleeping); the social-emotional quality of
caregiver-child interactions (such as presence of warm vs. harsh tones, or high vs. low synchrony between
caregiver and infant); and parent and child aect, including distress signals (such as fussing and crying) and
parental stress. Second, this platform should be able to collect such data for caregiver-child activity for at least a
full week or longer in their natural environment. While such extended recordings are still relatively rare, one
week of data could likely provide sucient variability in natural activity without being too burdensome for our
families.
The current paper is focused on the development of a model to automatically detect holding behaviors
via mother and infant motion signals. However, the additional requirements of our system aect the design
considerations and sensor selection of the current research and, as such, we provide additional context on our
overall platform.
Our current sensor platform includes high-quality heart rate and motion sensors worn by mother and child,
as well as a continuous audio recording device, worn by infants [
30
]. These devices were selected to provide
Proc. ACM Interact. Mob. Wearable Ubiquitous Technol., Vol. 3, No. 2, Article 64. Publication date: June 2019.
Automated Detection of Infant Holding Using Wearable Sensing: Implications for Developmental ... 64:5
access to various aspects of parent and child activity. In particular, body-worn motion sensors can ostensibly
be used to detect infant behavior and and caregiving activities (detailed in later paragraphs). Additionally, we
envision using audio recordings to automatically detect qualitative and aective aspects of the parent and child
interactions [
27
,
42
]. High-quality heart rate data provides an opportunity to examine physiological measures
related to social-emotional well-being and parent-child regulation, including vagal tone and parent-child heart
rate synchrony [
15
]. Additionally, heart rate data can improve the quality of collected sleep data, another key
behavior in our study. Given these varied goals, there were a number of competing considerations in choosing
the form factor for the motion sensor used in the current study. While a wrist-worn sensor would provide a more
comfortable experience for long-term wear, the need for high-quality heart rate data, as stated above, necessitates
a chest-worn sensor. Ultimately, we selected a single chest-worn sensor, collecting both motion and high-quality
heart rate signals, as our primary data collection device. However, we also recruited a small sample of participants
to wear wrist-worn motion sensors to assess the feasibility of this alternative placement.
In our ongoing data collection, families with infants aged six weeks to nine months participate in two sessions,
a 90-minute "Introductory" session (henceforth referred to as the Intro session) and a 72-hour "Home Recording"
session, in which they use our sensor platform to collect 72 hours of data over the course of a week. In the current
paper, we focus exclusively on data collected in the Intro session, which is detailed in Section 3.1 below. Intro
sessions are videotaped, thus providing us a highly reliable source of ground truth data to develop and assess
our holding detection models. We describe the detailed protocol for the Intro Session in Section 3, as well as the
accuracy of our models and their implications for future studies in Section 4and 5.
2.2 Assessment Scenarios
The literature suggests that both the duration and the timing of holding behaviors can impact infant outcomes
[
1
,
2
]. In light of this body of knowledge we dened ve relevant assessment scenarios (listed in Table 1). In
the simplest case, knowing the accurate onset and oset of individual holding events at second-by-second level
(Assessment Scenario 1) can be valuable for we can infer exact interaction patterns. Some hypotheses are more
specic with reference to timing: attachment theory suggests that knowing whether the parent is contingently
holding following a stimulus, such as crying, can determine its outcome. Thus, knowing the timing of holding
events would be valuable (Assessment Scenario 2). More broadly, we also want to know whether or not the parent
picks up the infant contingent to the onset of distress episode (Assessment Scenario 3). In addition, knowing how
much a parent is holding relative to other families (Assessment Scenario 4 & 5) would be valuable for predicting
outcomes such as attachment security or maternal mental health systems. As reported above, more daily carrying
can eectively reduce infant crying and fussing [
24
]. Each of these scenarios suggest a dierent way to conduct
automated activity recognition. Developmental scientists may use an automated assessment method to study
any of these scenarios (summarized in Table 1). Furthermore, dierences among the evaluation scenarios can be
considered through an automated assessment system. For example, while an algorithm may be able to perform
well on average for Assessment Scenario 1, it may not have individual dierences for Assessment Scenario 4 & 5.
So it is important to evaluate and iterate on the models to get to adequate performance for each one of these
scenarios. We will explore this in Section 4.1. These dierent scenarios are of value to developmental scientists
as well as more broadly to social scientists interested in leveraging activity recognition within studies of daily
activity.
3 METHOD
3.1 Data Collection
3.1.1 Protocol. The current study uses data collected within an Intro session lasting approximately 90 minutes.
Following an introduction to the study goals and the collection of informed consent, parents and their infants
Proc. ACM Interact. Mob. Wearable Ubiquitous Technol., Vol. 3, No. 2, Article 64. Publication date: June 2019.
64:6 X. Yao et al.
Table 1. Descriptions and Examples of Assessment Scenarios for Activity Recognition
Assessment Scenario Description Example
1. High Temporal Precision
Accurate onset and oset of indi-
vidual holding events
What is the timing of holding be-
haviors over the course of the
day? Is holding or holding dura-
tion contingent upon infant dis-
tress or is it unrelated to infant
activity?
2. Event-based Accuracy (with
Thresholds)
Successful identication of a spe-
cic activity obtaining a cer-
tain amount of overlap between
ground truth and prediction
How many times is the infant
held over the course of a day?
Is rate of holding decreased in
cases of maternal depression?
3. Contingency Analysis
Determine whether an activity
happens with a certain window
of time
Is infant held at all within two
minutes of crying onset?
4. Absolute Activity Summaries
Total sum of the amount of an
activity
How much time is a child held
over the course of a day?
5. Relative Activity Summaries
The amount of an activity com-
paring to others and dierent
days.
Does the model accurately iden-
tify the rank order of holding
time across children? Are chil-
dren who are held more often
than others accurately identied
by the model as such?
participated in developmental assessments and naturalistic activities while wearing our sensor suite (detailed in
Section 3.1.3), and each session was recorded on video. At the start of every video recording, research personnel
completed a sequence of movements with distinct motion and visual pattern to synchronize the video and sensor
data [36].
The Intro session tasks included two commonly-used developmental assessments: a 5-minute standardized
"free-play" task, in which caregivers were asked to play with their infants as they normally would, while seated
on the oor, and a 5-to-7-minute "reactivity assessment" [
26
] in which infants are presented with a number
of novel or "intense" stimuli (including diluted lemon juice and noisy toys) in a controlled manner to assess
individual dierences in their reactions. Additionally, in a nal activity named "around-the-house" task, caregivers
were asked to mimic typical care-giving routines in various locations around their own homes (10-20 minutes).
For example, we asked mothers to show us how they typically played with, changed, and fed their infants, in
the locations they would commonly do so. This task was developed specically to provide a variety of natural
physical contact and proximity behaviors, including numerous unprompted opportunities for parents to pick up,
Proc. ACM Interact. Mob. Wearable Ubiquitous Technol., Vol. 3, No. 2, Article 64. Publication date: June 2019.
Automated Detection of Infant Holding Using Wearable Sensing: Implications for Developmental ... 64:7
carry and hold their infants as they moved them around in their homes. Video recorded data from the entire
Intro session was included as training data in our holding recognition model.
3.1.2 Participants. 33 healthy mother-infant pairs were recruited to participate in our study. The average age
of the infants was 5.33 months (std: 2.54, range: 1.23-10.8), and the sample included 13 females and 20 males.
Maternal age was collected from all but two participants in our sample, and the average age of the mothers was
30.71 (n:31, std: 3.94, range: 23-39). 19 mothers identied as White, 5 as Hispanic, 2 as African-American, 1 as
Asian, and 6 as Multiracial.
3.1.3 Sensors. As detailed in Section 2, chest-worn sensors were used for our main data collection eorts,
due to requirements for high-quality heart rate data in a larger ongoing study (N=26). Additionally, to assess
the feasibility and trade-os with a likely more comfortable option for long-term wear, a smaller sample of
participants (N =12) wore wrist-worn sensors during the Intro session. Note that 5 of the 33 participants collected
both chest and wrist-worn data during their session.
We used the Movisens EcgMove 3 (Movisens) [
34
] to collect chest-worn sensor data. This sensor collects
electrocardiogram (ECG; up to 1024Hz), 3-axis acceleration (64 Hz), and barometric pressure (1 Hz) data. It can
collect continuous data for 3 days without charging and store up to 2 weeks of data. Our holding recognition
model relies on the acceleration data alone. The sensor was attached to both caregivers and infants using either a
chest belt or adhesive electrodes. Figure 1a shows an infant wearing a Movisens on the chest.
We used the MetaMotionR (MMR) [
32
] to collect wrist-worn sensor data. The MMR is a 9-axis IMU and
environmental sensor, but for purposes of comparison, only 3-axis accelerometer data was used in the current
study. It collects movement data at 25 Hz. Caregivers wore the MMR sensor on their dominant wrist; infants
wore the MMR attached to their ankle, as shown in Figure 1b.
3.1.4 Annotation. Trained coders annotated various states of physical contact (e.g. bouncing, carrying) occurring
throughout the entire duration of the Intro session. Proximity states were collapsed into a binary category of
holding and not holding (Mean: 45 minutes, SD: 11 minutes, kappa = 0.844). Holding activity includes both
holding and carrying where a mother is physically supporting her infant while not holding mainly relates to
those activities where a mother is away from her infant. The detailed denition for holding and not holding
can be found in Table 2. In this way, we obtained ne-grained distinctions of ground truth at second-by-second
resolution. We estimated that during a session, mothers held their infants on average 41.4% of the time (std: 0.130,
range: 0.163 - 0.744), which was about 18.6 minutes for an average 45-minute session.
Table 2. Definitions of holding activities between mothers and infants
Activities Denition
Holding
Caregiver is physically supporting the infant while either standing or
sitting. Includes carrying, picking up, putting down and bouncing so
long as child’s weight is completely supported by caregiver.
Not holding
Caregiver is not supporting the weight of the baby. Includes all periods
without physical contact as well as touching or leaning behaviors in
which physical contact does not completely support the child’s weight.
Proc. ACM Interact. Mob. Wearable Ubiquitous Technol., Vol. 3, No. 2, Article 64. Publication date: June 2019.
64:8 X. Yao et al.
(a) (b)
Fig. 1. Images of sensors as worn by participants in our study. Chest-worn sensors were aached with a chest belt or adhesive
electrodes as in Figure 1a. Wrist-worn sensors were aached to mothers’ dominant wrist or infants’ ankle with tight-fiing
fabric cus (shown in pink), as indicated in 1b.
3.2 Development of Holding Detection Algorithm
Our multi-stage algorithm has been designed for automatic holding detection and contains four main steps: i)
preprocessing and windowing (to extract small, consecutive portions of sensor data for processing); ii) feature
extraction; iii) machine learning based classication; and iv) post-processing of prediction results through
smoothing. We applied the algorithm to 3-axis acceleration data streams recorded using the sensing system
mentioned in Section 3.1.3 (separately for chest-worn and wrist-worn setting). Model validation and evaluation is
based on a leave-one-participant-out protocol.
Preprocessing calculates the magnitude of raw acceleration data for both mother and infant, and then smoothes
the resulting sensor data stream using a standard Savitzky-Golay lter. Sliding window based analysis employs a
7-second analysis window that is shifted in increments of 1 second. For each window we compute ve features:
i) correlation of acceleration between mother and infant; ii) squared correlation; iii) variance of mother’s
acceleration; iv) variance of baby’s acceleration; and v) the dierence between variances. The correlation features
were chosen to capture synchrony between caregiver and infant when holding or carrying, because in those cases
their data streams are expected to be similar, which is in contrast to when both are moving independently. We
expected to see high-correlation patterns during holding and little to none correlation during not holding. Variance
features were chosen to characterize the individual variations in movements. Small dierences suggest similar
movement distributions, whereas larger dierences suggest very dierent movement patterns. The machine
learning based classication, which is derived from supervised training, then assigns activity labels to each
window, and nal predictions are determined through majority vote on predictions for overlapping windows.
We explored four machine learning models: i) AdaBoost; ii) Logistic Regression; iii) Random Forest; and iv)
Support Vector Machines (SVM). Random Forest provided the highest and the most stable accuracy across all
sessions and therefore we limit the presentation of the results in Section 4. to that classier.
A nal post-processing step involved smoothing the stream of predictions to remove implausible outliers in the
predictions that would suggest unrealistically short holding (or not-holding) episodes. In our model, non-holding
Proc. ACM Interact. Mob. Wearable Ubiquitous Technol., Vol. 3, No. 2, Article 64. Publication date: June 2019.
Automated Detection of Infant Holding Using Wearable Sensing: Implications for Developmental ... 64:9
episodes that were shorter than or equal to 30 seconds, and holding episodes that were shorter than 10 seconds
were eliminated by the smoothing procedure through reassignment to the particular other class.
4 RESULTS
As motivated in Section 2, we introduced 5 assessment scenarios (Table 1) which are necessary to consider
in social science studies. We will show the results derived from our model and chest-worn Movisens sensor
one-by-one here and discuss its implications in Section 5. In addition, we present the results for all 5 scenarios
using data collected by wrist-worn sensors and compare it to chest-worn sensors.
4.1 Assessment Scenario 1: High Temporal Precisions
Assessment Scenario 1 represents conventional accuracy metrics used in activity recognition. It classied holding
vs. not holding activities at a second-by-second level. This level of granularity (knowing what the participant is
doing every second) is important to infer exact interaction patterns.
Model performance for classifying holding and not holding second reached 87% of accuracy and 83.1% of F1
score as in Table 3. Precision and recall were both higher than 80%, meaning our model is of practical relevance
for correctly distinguishing holding from not holding episodes.
Table 3. Second-by-second Accuracy
Accuracy F1 Average Holding precision Holding recall
Mean 0.870 0.831 0.818 0.854
Std 0.059 0.091 0.105 0.102
4.2 Assessment Scenario 2: Event-based Accuracy
Our second scenario assesses model performance for identifying events, that is, periods of continuous holding
between periods of not-holding. To assess event-related accuracy, we rst assessed event recall, that is, how often
the model accurately detected any frames of holding during a holding event identied in the ground truth data.
Additionally, we set dierent thresholds to specify varying proportions of overlap between detected and ground
truth holding events, which could theoretically range from 0% (no detected holding for a given ground truth
holding event) to 100% (complete overlap between a detected holding event and the corresponding ground truth
event). We present mean recall and mean precision (summarized in Figure 2) across overlap thresholds from
0%-80%. Our results indicate that we can obtain recall and precision larger than 75%, with a threshold of 60% of
overlap matching between ground truth and predicted events. In particular, at the 60% threshold, the mean recall
was 0.828 and mean precision 0.774. However, this threshold also results in a number of false positives. Where on
average the model detected 9.308 (Std: 3.739) events, 2.077 (Std: 1.639) were false positives.
Visual inspections suggested that these false positives were more likely to occur when ground truth events
were very short. Considering only those events that were 15 seconds or greater, the number of false positives
was reduced. In particular, an average number of 1.615 (Std: 1.003) false positives was generated for an average
session with 6.962 (Std: 1.951) holding events.
These results indicate that while our model does not accurately capture the entire duration of every holding
event, it can be used to calculate the frequency of holding events, especially those longer than 15 seconds. Measures
of holding frequency have previously been used to assess variation in physical contact in the developmental
literature [39].
Proc. ACM Interact. Mob. Wearable Ubiquitous Technol., Vol. 3, No. 2, Article 64. Publication date: June 2019.
64:10 X. Yao et al.
Fig. 2. Mean recall and precision values of all 26 sessions for assessment scenario 2. When all events are considered, the
model can capture at most 88% of the holding events and the number was increased to 95% as short holding events without
a strong paern are missed. The mean precision didn’t change much for dierent overlap thresholds overall.
4.3 Assessment Scenario 3: Contingency Analysis
The timing of caregiving behaviors is of special interest to developmental psychologists. In particular, caregiving
behaviors which are contingent upon children’s activities have been shown to promote learning and development
across a number of domains, from language learning to attachment and social-emotional outcomes [
5
,
7
,
25
].
Parental holding contingent upon infant distress is one input considered to be important for development of
attachment behaviors [
24
]. Thus in our third scenario we assessed the accuracy of our model to detect the
presence or absence of any holding activity within specic windows of time. In future research eorts these
windows could be determined via the onset of periods of infant distress.
We considered windows of 5 seconds (one second stride) and 2 minutes (one minute stride) in length. If the
model predicted any holding events within the window, the whole window was labeled as "holding", otherwise it
was labeled as "not holding". We chose two dierent window lengths in consideration of two cases common in
home scenarios. First, mother can be feet or rooms away from the baby or she may be busy with another child,
which makes window length of 2 minutes an appropriate metric, whereas window length of 5 seconds is for cases
when a caregiver is next to the infant, typical during interactions such as home play.
Our results are reported in the form of confusion matrices in Table 4, Considering 5-second windows (13,312
windows total) summed across all 26 sessions, the precision for holding activities reached 0.850 while the recall
was 0.873 (Accuracy: 0.881, F1-score: 0.861). Considering 2-minute windows (1,097 windows total) summed across
Proc. ACM Interact. Mob. Wearable Ubiquitous Technol., Vol. 3, No. 2, Article 64. Publication date: June 2019.
Automated Detection of Infant Holding Using Wearable Sensing: Implications for Developmental ... 64:11
all 26 sessions, the precision for holding activities reached 0.843 while the recall was 0.943 (Accuracy: 0.847,
F1-score: 0.891).
Results considering each participant separately (using a window size of 2 minutes) are reported in Table 5.
The high values in all 4 measures across all sessions mean that our model can identify contingencies with high
condence.
Table 4. Confusion Matrices of Contingency-Analysis Assessment (Scenario 3) for Windows Summed across All 26 Sessions
(a) 5-sec Confusion Matrix
Ground
Truth
Predictions
Not Holding Holding
Not Holding 6801 870
Holding 717 4924
(b) 2-min Confusion Matrix
Ground
Truth
Predictions
Not Holding Holding
Not Holding 245 127
Holding 41 684
Table 5. Performance of Contingency-Analysis Assessment (Scenario 3) for Each Participant (Considering 2-min Windows)
Session Accuracy F1 average Hold Precision Hold Recall
Mean 0.855 0.894 0.855 0.944
Std 0.081 0.067 0.088 0.069
4.4 Assessment Scenario 4 & 5: Absolute and Relative Activity Summaries
Our nal assessment scenarios characterize relative dierences in summed activity between individuals (or in
our case, families). A key goal within developmental science is to understand how dierences in the quantity
or quality of behaviors between individuals or families predict children’s future developmental outcomes. This
requires condence in the model’s predictions for each pair of participants rather than average model performance
across the entire sample. We note that it is common for model performance to vary across participants: in our
assessment of second-by-second accuracies, the standard deviation of our model’s recall was larger than 10% (see
Section 4.1), which is not uncommon in published activity recognition models. For this reason we developed
two additional scenarios to directly test the model performance to preserve absolute and relative dierences in
activity summaries between participants.
For the fourth assessment scenario, we compared ground truth and predicted holding duration during each
session using Pearson’s correlation. Due to slightly varying session lengths we calculated relative holding
durations after normalizing session length to 45 minutes. The R2 of the correlation was strong (correlation =
0.915, R2 = 0.815; Figure 3Left Plot), indicating that our model is accurate in predicting the amount of holding
across our participants.
For the fth assessment scenario, we assessed whether our model could preserve the rank of the amount of
holding, which in some cases may be preferable to absolute dierences between individuals. Holding time was
again normalized to adjust for dierences in recording durations between participants and compared between
ground truth and predictions. The right plot of Figure 3shows the rank of holding time in both ground truth
and predictions and all data points are close to the line of expected output. Thus, our model preserved most of
the order and can be used to assess the individual dierence in the amount of holding. The Spearman’s rank
correlation reaches 0.876 and R2 is 0.771.
Proc. ACM Interact. Mob. Wearable Ubiquitous Technol., Vol. 3, No. 2, Article 64. Publication date: June 2019.
64:12 X. Yao et al.
(a) Duration of Holding Time (R-squared = 0.815) (b) Rank of Holding Time (R-squared = 0.767)
Fig. 3. These plots shows the correlations between ground truth and predictions. In particular, the le plot corresponds
to Scenario 4 and right plot Scenario 5. Each individual point represents one session. For the le plot, x-axis represents
the holding time normalized to a 45-minute session in ground truth for each individual session and y-axis represents the
predicted holding time. For the right plot, x-axis represents the rank for the amount of holding time in ground truth and
y-axis represents the predicted rank. In the ideal situation, we should obtain a diagonal line where the predicted rank is the
same as the ground truth rank.
4.5 Comparison between Chest-worn and Wrist-worn Sensor
In our nal analyses we compare the accuracy of models developed using data from wrist-worn sensors with our
original model developed using data from chest-worn sensors. As stated above, this comparison has implications
for participant usability as wrist-worn sensors are known to be more comfortable for long-term use. To develop
the model from the wrist-worn data we used the same methods described in Section 3.2 using 12 total sessions.
Overall model accuracy for the 12 wrist-worn sessions was 0.738 (std: 0.076, F1: 0.690, Precision: 0.744, Recall:
0.679) for Scenario 1, with similar accuracies for the other 4 scenarios. Additionally, ve participants wore wrist
and chest-worn sensors systems simultaneously, allowing us to directly compare their results for each of the ve
assessment scenarios. These are summarized in Table 6.
The data in Table 6indicates that wrist-worn sensors show worse classication of holding than chest-worn
sensors. Interestingly, these discrepancies in classication accuracy are not consistent throughout all ve scenarios.
While results in scenarios 2 and 5 appear inadequate, wrist-worn sensing is appears adequate for use cases
according to scenarios 3 and 4.
Proc. ACM Interact. Mob. Wearable Ubiquitous Technol., Vol. 3, No. 2, Article 64. Publication date: June 2019.
Automated Detection of Infant Holding Using Wearable Sensing: Implications for Developmental ... 64:13
Table 6. Comparison of accuracy between wrist-worn sensors and chest-worn sensors in all five scenarios (recorded simulta-
neously in the same five sessions). Note that for Scenario 2, precision and recall were calculated for all events longer than 15
seconds with over 60% of overlap. In addition, scenario 3 is the mean measurements of 2 minute contingency windows.
Wrist-worn Chest-worn
Scenario 1 Precision Recall Precision Recall
0.784 0.716 0.826 0.868
Scenario 2 Precision Recall Precision Recall
0.693 0.709 0.859 0.949
Scenario 3 Precision Recall Precision Recall
0.84 0.919 0.862 0.955
Scenario 4 Correlation R2 Correlation R2
0.878 0.747 0.952 0.878
Scenario 5 Correlation R2 Correlation R2
0.600 0.360 0.800 0.640
5 DISCUSSION
5.1 Implications
Our results indicate that high precision and recall were achieved for all assessment scenarios, meaning that our
model can automatically detect holding activities accurately and consistently within and across all participants
in our study.
The strong performance of our model means that it can be used to precisely and objectively quantify holding
behaviors in naturalistic settings, useful for both basic science of child development as well as development
of interventions. For example, data on absolute or relative amounts of holding across a day could be used to
predict relevant child and caregiver outcomes ranging from stress system neurobiology and attachment security
to maternal mental health symptoms. The temporal precision of the model paired with its accuracy across
participants means it could be used to collect data speaking to specic hypotheses regarding the role of the
timing of physically soothing responses to infant distress. Research indicates that holding can be an important
mechanism to reduce infant crying and promote attachment behaviors fundamental to lifelong outcomes in social
and emotional domains.
Our model could be incorporated into interventions designed to increase caregiver holding behaviors. Providing
families with objective feedback about their holding behaviors could function to increase holding frequencies and
durations, with benets for both caregivers and infants. The LENA system [
30
] is a similar intervention model
that has had much success in the developmental science community. LENA is a wearable audio recorder which
automatically detects patterns of contingent speech between parents and their children. This system has been
used by thousands of parents to provide objective feedback used to increase speaking to children and ultimately
vocabulary and school success.
Finally, our assessment scenarios could be usefully applied to many dierent activity recognition problems.
In particular, we highlight the need to directly measure the consistency of model performance across dierent
scenarios. In particular, models with acceptably high average accuracy may obscure or mislead researchers
interested in using activity recognition to identify individual dierences across participants [
11
]. Assessment and
reporting of accuracy across participants or sessions can lead to more consistent models or at the very least more
informed use of these models.
Proc. ACM Interact. Mob. Wearable Ubiquitous Technol., Vol. 3, No. 2, Article 64. Publication date: June 2019.
64:14 X. Yao et al.
5.2 Assessment of Sensor Usability
One potential concern with our sensing platform is that it may be physically uncomfortable for our participants.
Attaching a sensor to the chest of an infant may limit their natural movements or restrict interactions with the
caregiver. Additionally it may simply be uncomfortable to wear for extended periods of time. To assess these
potential concerns, we conducted an exit survey among a subset of our participants (N = 18) to evaluate their
experiences with our current sensing platform. The survey was conducted following the home recording session,
in which participants were instructed to wear our complete sensor platform for 72 hours over the course of a
single week.
On a scale of 1 to 5, we asked caregivers to indicate how comfortable they were with the sensor attached to the
chest, where 1 represents "very uncomfortable" and 5 "very comfortable". We achieved an average score of 3.67,
which is reassuring for the developed approach. In particular, 10 caregivers said that the chest-worn sensor was
comfortable/very comfortable while only one participant said the sensor was uncomfortable. In addition, using the
same scale, 17 (out of 18) mothers provided ratings of the perceived comfort of their infants. Overall, mothers
perceived the infant to be comfortable with the chest-worn sensors, with an average rating of 4.12. Only one
mother reported that she perceived her infant was uncomfortable with the sensors while 13 mothers reported
that their infants were either comfortable or very comfortable.
Overall, our survey results indicate that a chest-worn sensor was suciently comfortable for our participants
and thus does not represent a reason for concern at this time.
5.3 Limitations and Future Work
Relative to wrist-worn sensors, chest-worn sensors provided higher accuracy across all assessment scenarios.
While our participants of our exit survey indicated that chest-worn sensors were suciently comfortable, (Section
5.2) 14 of 18 respondents indicated that they would prefer to use wrist-worn sensors over chest-worn sensors if
provided the option. Thus, there is an apparent trade-o between sensor positioning (and therefore accuracy of
holding detection) and participant comfort. As a consequence, in future work we will continue to optimize our
sensing platform. We note that prototypes of miniaturized inertial measurement units have recently become
available. These prototypes resemble a band-aid like form factor and are thus substantially smaller than the
devices used for the study presented in this paper [
33
]. This prospective form factor could provide a more
comfortable user experience while maintaining the higher accuracy of chest-worn sensing, eectively eliminating
any remaining concerns.
While our model showed reliable performance in detecting the presence or absence of holding behavior, we
recognize that dierent types of physical contact may have dierent implications for children’s immediate and
long term outcomes. For example, in studies of infant massage, babies who received moderate pressure had
better gastric motility and vagal tone than those who received light pressure [
12
]. It is likely that other types
of touch, such as bouncing or carrying versus a more procedural contact (e.g., fastening or adjusting clothing)
also have diering implications for attachment and stress system development across the rst year. We are not
aware of any research dierentiating these in the developmental science literature, likely because it has been
impossible to accurately characterize or report upon such variations in touch in natural daily activity. While
beyond the scope of this paper, more precise characterization of holding activities may be possible with future
research eorts. In particular, in our current dataset we have annotated a total of eight distinct types of physical
contact, including: holding, carrying, bouncing, picking up, putting down, hovering, touching and nearby. We
will work to incorporate these into our framework in order to provide more ne-grained analysis of physical
contact behaviors in the future.
Proc. ACM Interact. Mob. Wearable Ubiquitous Technol., Vol. 3, No. 2, Article 64. Publication date: June 2019.
Automated Detection of Infant Holding Using Wearable Sensing: Implications for Developmental ... 64:15
5.4 Conclusions
In the current paper, we described our use of wearable sensors to build an overarching system to capture multiple
aspects of caregiver-infant interaction. We presented a model that combines chest-worn accelerometer data from
caregiver and infant to detect holding behavior. Our model allows objective assessments of the amount and
duration of holding in naturalistic home settings, with implications for basic science and intervention. Our results
show that our model achieves high average precision and recall at second-by-second resolution. In addition, we
proposed and discussed accuracy metrics for four other potential assessment scenarios, namely, event-based
accuracy, contingency analysis, absolute and relative activity summaries. A subset of our 26 total participants
wore both wrist- and chest-worn sensors, allowing us to compare accuracy between these two form factors. Our
results indicate that chest-worn sensors are more accurate while still reasonably comfortable, even in deployments
lasting up to a week. However, wrist-worn sensors may be preferable for longer-term applications when only
summary measures of holding time are desired.
REFERENCES
[1]
Mary D. Salter Ainsworth. 1979. Attachment as Related to Mother-Infant Interaction. Advances in the Study of Behavior, Vol. 9.
Academic Press, 1 – 51. https://doi.org/10.1016/S0065-3454(08)60032-7
[2]
Elizabeth Anisfeld, Virginia Casper, Molly Nozyce, and Nicholas Cunningham. 1990. Does infant carrying promote attachment? An
experimental study of the eects of increased physical contact on the development of attachment. Child development 61, 5 (Oct. 1990),
1617–1627. https://doi.org/10.2307/1130769
[3]
Akin Avci, Stephen Bosch, Mihai Marin-Perianu, Raluca Marin-Perianu, and Paul Havinga. 2010. Activity Recognition Using Inertial
Sensing for Healthcare, Wellbeing and Sports Applications: A Survey. In 23th International Conference on Architecture of Computing
Systems 2010. 1–10.
[4]
Theodore Beauchaine. 2001. Vagal tone, development, and Grays motivational theory: Toward an integrated model of autonomic
nervous system functioning in psychopathology. Development and Psychopathology 13, 2 (Feb. 2001), 183–214. https://doi.org/10.1017/
s0954579401002012
[5]
Beatrice Beebe, Joseph Jae, Sara Markese, Karen Buck, Henian Chen, Patricia Cohen, Lorraine Bahrick, Howard Andrews, and Stanley
Feldstein. 2010. The origins of 12-month attachment: A microanalysis of 4-month mother-infant interaction. Attachment & Human
Development 12, 1-2 (2010), 3–141. https://doi.org/10.1080/14616730903338985
[6]
Bettina Bohnhorst, Tim Heyne, Corinna S. Peter, and Christian F. Poets. 2001. Skin-to-skin (Kangaroo) care, respiratory control, and
thermoregulation. The Journal of pediatrics 138, 2 (Feb. 2001), 193–197. https://doi.org/10.1067/mpd.2001.110978
[7]
Marc H. Bornstein and Catherine S. Tamis-Lemonda. 1997. Maternal responsiveness and infant mental abilities: Specic predictive
relations. Infant Behavior and Development 20, 3 (1997), 283 – 296. https://doi.org/10.1016/S0163- 6383(97)90001-1
[8]
Ksenia Bystrova, Valentina Ivanova, Maigun Edhborg, Ann-So Matthiesen, Anna-Berit Ransjö-Arvidson, Rifkat Mukhamedrakhimov,
Kerstin Uvnäs-Moberg, and Ann-Marie Widström. 2009. Early Contact versus Separation: Eects on Mother-Infant Interaction One Year
Later. Birth 36, 2 (June 2009), 97–109. https://doi.org/10.1111/j.1523-536X.2009.00307.x
[9]
Agustin Conde-Agudelo, José M. Belizán, and Jose Lius Diaz-Rossello. 2016. Kangaroo mother care to reduce morbidity and mortality in
low birthweight infants. Cochrane Database of Systematic Reviews 23, 8 (Aug. 2016). https://doi.org/10.1002/14651858.CD002771.pub4
[10]
Louis J. Cozolino. 2014. The neuroscience of human relationships: attachment and the developing social brain. W. W. Norton & Company,
New York, NY.
[11]
Kaya de Barbaro. 2019. Automated sensing of daily activity: A new lens into development. Developmental Psychobiology (March 2019),
1–21. https://doi.org/10.1002/dev.21831
[12]
Miguel A. Diego, Tiany Field, and Maria Hernandez-Reif. 2005. Vagal Activity, Gastric Motility, and Weight Gain in Massaged Preterm
Neonates. The Journal of pediatrics 147, 1 (July 2005), 50–55. https://doi.org/10.1016/j.jpeds.2005.02.023
[13]
Byron Egeland and Ellen A. Farber. 1984. Infant-Mother Attachment: Factors Related to Its Development and Changes over Time. Child
Development 55, 3 (June 1984), 753–771. https://doi.org/10.2307/1130127
[14]
Mingming Fan, Dana Gravem, Dan M. Cooper, and Donald J. Patterson. 2012. Augmenting Gesture Recognition with Erlang-cox Models
to Identify Neurological Disorders in Premature Babies. In Proceedings of the 2012 ACM Conference on Ubiquitous Computing (UbiComp
’12). ACM, New York, NY, USA, 411–420. https://doi.org/10.1145/2370216.2370278
[15]
Ruth Feldman, Romi Magori-Cohen, Giora Galili, Magi Singer, and Yoram Louzoun. 2011. Mother and infant coordinate heart rhythms
through episodes of interaction synchrony. Infant Behavior and Development 34, 4 (Dec. 2011), 569–577. https://doi.org/10.1016/j.infbeh.
2011.06.008
[16] Tiany Field. 1995. Touch in Early Development. Lawrence Erlbaum Associates, Mahwah, NJ.
Proc. ACM Interact. Mob. Wearable Ubiquitous Technol., Vol. 3, No. 2, Article 64. Publication date: June 2019.
64:16 X. Yao et al.
[17]
Tiany Field, Miguel Diego, and Maria Hernandez-Reif. 2010. Preterm Infant Massage Therapy Research: A Review. Infant behavior and
development 33, 2 (April 2010), 115–124. https://doi.org/10.1016/j.inf beh.2009.12.004
[18]
Larry Gray, Lisa B. Watt, and Elliott Blass. 2000. Skin-to-Skin Contact Is Analgesic in Healthy Newborns. Pediatrics 105, 1 (Jan. 2000),
e14. https://doi.org/10.1542/peds.105.1.e14
[19]
Nils Y. Hammerla, James M. Fisher, Peter Andras, Lynn Rochester, Richard Walker, and Thomas Plotz. 2015. PD Disease State Assessment
in Naturalistic Environments Using Deep Learning. In Proceedings of the Twenty-Ninth AAAI Conference on Articial Intelligence
(AAAI’15). AAAI Press, 1742–1748.
[20]
Gillian R. Hayes, Karen G. Cheng, Sen H. Hirano, Karen P. Tang, Marni S. Nagel, and Dianne E. Baker. 2014. Estrellita: A Mobile Capture
and Access Tool for the Support of Preterm Infants and Their Caregivers. ACM Trans. Comput.-Hum. Interact. 21, 3, Article 19 (June
2014), 28 pages. https://doi.org/10.1145/2617574
[21]
Matthew J. Hertenstein. 2002. Touch: Its Communicative Functions in Infancy. Human Development 45, 2 (2002), 70–94. https:
//doi.org/10.1159/000048154
[22]
Matthew J. Hertenstein, Julie M. Verkamp, Alyssa M. Kerestes, and Rachel M. Holmes. 2006. The Communicative Functions of Touch in
Humans, Nonhuman Primates, and Rats: A Review and Synthesis of the Empirical Research. Genetic, social, and general psychology
monographs 132, 1 (Feb. 2006), 5–94. https://doi.org/10.3200/MONO.132.1.5-94
[23]
Jesse Hoey, Thomas Plötz, Dan Jackson, Andrew Monk, Cuong Pham, and Patrick Olivier. 2011. Rapid Specication and Automated
Generation of Prompting Systems to Assist People with Dementia. Pervasive Mob. Comput. 7, 3 (June 2011), 299–318. https://doi.org/10.
1016/j.pmcj.2010.11.007
[24]
UA Hunziker and Ronald Barr. 1986. Increased carrying reduces infant crying: A randomized controlled trial. Pediatrics 77, 5 (June
1986), 641–648.
[25]
Joseph Jae, Beatrice Beebe, Stanley Feldstein, Cynthia L. Crown, Michael D. Jasnow, Philippe Rochat, and Daniel N. Stern. 2011.
Rhythms of dialogue in infancy: Coordinated timing in development. Monographs of the Society for Research in Child Development 66, 2
(2011), i–viii+1–149.
[26]
Jerome Kagan, J Steven Reznick, and Nancy Snidman. 1987. The Physiology and Psychology of Behavioral Inhibition in Children. Child
development 58, 6 (Dec. 1987), 1459–1473. https://doi.org/10.2307/1130685
[27]
Jonathan C. Kim and Mark A. Clements. 2015. Formant-based feature extraction for emotion classication from speech. In 2015 38th
International Conference on Telecommunications and Signal Processing (TSP). 477–481. https://doi.org/10.1109/TSP.2015.7296308
[28]
Marshall H. Klaus, Richard Jerauld, Nancy C. Kreger, Willie McAlpine, Meredith Stea, and John H. Kennell. 1972. Maternal Attachment.
New England Journal of Medicine 286, 9 (Feb. 1972), 460–463. https://doi.org/10.1056/NEJM197203022860904
[29]
Matthias Kranz, Andreas Möller, Nils Hammerla, Stefan Diewald, Thomas Plötz, Patrick Olivier, and Luis Roalter. 2013. The mobile
tness coach: Towards individualized skill assessment using personalized mobile devices. Pervasive and Mobile Computing 9, 2 (2013),
203 – 215. https://doi.org/10.1016/j.pmcj.2012.06.002 Special Section: Mobile Interactions with the Real World.
[30] LENA. [n. d.]. LENA. Retrieved April 2, 2019 from https://www.lena.org/
[31]
Emily Elysia Little, Cristine H. Legare, and Leslie J. Carver. 2018. Mother-Infant Physical Contact Predicts Responsive Feeding among
U.S. Breastfeeding Mothers. Nutrients 10, 9, Article 1251 (2018). https://doi.org/10.3390/nu10091251
[32] MbientLab. [n. d.]. MMR. Retrieved April 2, 2019 from https://mbientlab.com/metamotionr/
[33]
Ross Miller. 2015. The craziest tness tracker ever is basically a smart band-aid. https://www.theverge.com/2015/1/4/7492199/
ces-2015- craziest-tness- tracker-bandaid-ampstrip
[34]
movisens. [n. d.]. ECG and Activity Sensor - EcgMove 3. Retrieved April 2, 2019 from https://www.movisens.com/en/products/
ecg-and- activity-sensor- ecgmove-3/
[35]
Yunyoung Nam and Jung Wook Park. 2013. Child Activity Recognition Based on Cooperative Fusion Model of a Triaxial Accelerometer
and a Barometric Pressure Sensor. IEEE Journal of Biomedical and Health Informatics 17, 2 (March 2013), 420–426. https://doi.org/10.
1109/JBHI.2012.2235075
[36]
Thomas Plötz, Chen Chen, Nils Y. Hammerla, and Gregory D. Abowd. 2012. Automatic Synchronization of Wearable Sensors and
Video-Cameras for Ground Truth Annotation – A Practical Approach. In Proceedings of the 2012 16th Annual International Symposium on
Wearable Computers (ISWC) (ISWC ’12). IEEE Computer Society, Washington, DC, USA, 100–103. https://doi.org/10.1109/ISWC.2012.15
[37]
Thomas Plötz, Nils Y. Hammerla, Agata Rozga, Andrea Reavis, Nathan Call, and Gregory D. Abowd. 2012. Automatic Assessment of
Problem Behavior in Individuals with Developmental Disabilities. In Proceedings of the 2012 ACM Conference on Ubiquitous Computing
(UbiComp ’12). ACM, New York, NY, USA, 391–400. https://doi.org/10.1145/2370216.2370276
[38]
Thomas Plötz, Paula Moynihan, Cuong Pham, and Patrick Olivier. 2011. Activity Recognition and Healthier Food Preparation. Atlantis
Press, Paris, 313–329. https://doi.org/10.2991/978-94-91216- 05-3_14
[39]
Jaipaul L. Roopnarine, Enayet Talukder, Deepa Jain, Priti Joshi, and Parul Srivastav. 1990. Characteristics of holding, patterns of
play, and social behaviors between parents and infants in New Delhi, India. Developmental Psychology 26, 4 (1990), 667–673. https:
//doi.org/10.1037/0012-1649.26.4.667
Proc. ACM Interact. Mob. Wearable Ubiquitous Technol., Vol. 3, No. 2, Article 64. Publication date: June 2019.
Automated Detection of Infant Holding Using Wearable Sensing: Implications for Developmental ... 64:17
[40]
L. Alan Sroufe. 1979. The coherence of individual development: Early care, attachment, and subsequent developmental issues. American
Psychologist 34 (Oct. 1979), 834–841. https://doi.org/10.1037/0003-066X.34.10.834
[41]
L. Alan Sroufe, Byron Egeland, Elizabeth A. Carlson, and W. Andrew Collins. 2005. The development of the person: The Minnesota study
of risk and adaptation from birth to adulthood (1st. ed.). The Guilford Press, New York, NY.
[42]
Erin Tobin, Heidi Kane, Daniel J. Saleh, Derek Wildman, Elizabeth Crabb Breen, Elizabeth Secord, and Richard Slatcher. 2015. Asthma-
Related Immune Responses in Youth With Asthma: Associations With Maternal Responsiveness and Expressions of Positive and Negative
Aect in Daily Life. Psychosomatic medicine 77, 8 (Oct. 2015), 892–902. https://doi.org/10.1097/psy.0000000000000236
[43]
Junqing Wang, Aisling Ann O’Kane, Nikki Newhouse, Geraint Rhys Sethu-Jones, and Kaya de Barbaro. 2017. Quantied Baby:
Parenting and the Use of a Baby Wearable in the Wild. Proc. ACM Hum.-Comput. Interact. 1, CSCW, Article 108 (Dec. 2017), 19 pages.
https://doi.org/10.1145/3134743
Received November 2018; revised April 2019; accepted April 2019
Proc. ACM Interact. Mob. Wearable Ubiquitous Technol., Vol. 3, No. 2, Article 64. Publication date: June 2019.
... Greenspan et al. (2021) estimated body position angle using pitch angle cut-points from a single sensor embedded in a garment in 3-month-olds. Yao et al. (2019) used a pair of sensors, one worn by the infant and one worn by the caregiver, to train machine learning models that were able to accurately classify the time infants spent held by caregivers. Notably, the Yao et al. study validated their method "in the wild" by collecting data in the home rather than relying only on a laboratory sample, which suggests the feasibility of this method for our proposed application. ...
... Although recent work provides an encouraging outlook for measuring body position in infants (Yao et al., 2019;Airaksinen et al., 2020;Greenspan et al., 2021), there are several open questions. First, because past studies of body position (Airaksinen et al., 2020;Greenspan et al., 2021) did not include caregivers holding infants as a category, it is unknown whether our proposed body position categories-prone, supine, sitting, upright, and held by caregiver-can be accurately classified. ...
... We collected synchronized video and inertial sensor data while infants were in different body positions, and used those data to train classifiers and then validate them against the gold standard (human coding from video observation). As in past work (Nam and Park, 2013;Yao et al., 2019;Airaksinen et al., 2020), our aim was to determine whether the overall accuracy of classification was high (> 90% of agreement between model predictions and ground truth data). Moreover, we assessed whether the method could accurately detect individual differences in how much time infants spend in different body positions, which is relevant for characterizing everyday motor experiences and their potential downstream effects on other areas of development (e.g., Soska et al., 2010;Oudgenoeg-Paz et al., 2012;Walle and Campos, 2014). ...
Article
Full-text available
How can researchers best measure infants' motor experiences in the home? Body position—whether infants are held, supine, prone, sitting, or upright—is an important developmental experience. However, the standard way of measuring infant body position, video recording by an experimenter in the home, can only capture short instances, may bias measurements, and conflicts with physical distancing guidelines resulting from the COVID-19 pandemic. Here, we introduce and validate an alternative method that uses machine learning algorithms to classify infants' body position from a set of wearable inertial sensors. A laboratory study of 15 infants demonstrated that the method was sufficiently accurate to measure individual differences in the time that infants spent in each body position. Two case studies showed the feasibility of applying this method to testing infants in the home using a contactless equipment drop-off procedure.
... Even though in our study we observed significant variability in parental touch (as indexed for instance by the distribution of the PICTS scores), it is probable that the measures we used were not sensitive to the extreme ends of the caregiver behaviour spectrum. Emerging technologies, such as devices recording body contact (Yao et al., 2019), could partially address this issue by allowing us to capture touching behaviors over extended periods of time and in infants' natural environment, and thus might be the future of touch research in infancy. Moreover, one of our main measures of interest, salivary oxytocin, has been associated with some controversies about its validity and specificity (Uvnäs-Moberg et al., 2020), and it has also yielded a substantial amount of missing data in our study. ...
Article
Full-text available
Caregiver touch is crucial for infants' healthy development, but its role in shaping infant cognition has been relatively understudied. In particular, despite strong premises to hypothesize its function in directing infant attention to social information, little empirical evidence exists on the topic. In this study, we investigated the associations between naturally occurring variation in caregiver touch and infant social attention in a group of 6-to 13-month-old infants (n = 71). Additionally, we measured infant salivary oxytocin as a possible mediator of the effects of touch on infant social attention. The hypothesized effects were investigated both short term, with respect to touch observed during parent-infant interactions in the lab, and long term, with respect to parent-reported patterns of everyday touching behaviors. We did not find evidence that caregiver touch predicts infant social attention or salivary oxytocin levels, short term or long term. However, we found that salivary oxytocin predicted infant preferential attention to faces relative to nonsocial objects, measured in an eye-tracking task. Our findings confirm the involvement of oxytocin in social orienting in infancy, but raise questions regarding the possible environmental factors influencing the infant oxytocin system.
... As an important research topic in ubiquitous computing, Human Activity Recognition (HAR) from wearable sensors is a key element of various intelligent applications, such as smart personal assistants [1], healthcare assessment [2][3][4][5][6], sports monitoring [7], and aging care [8]. HAR can recognize a variety of individual behaviors, such as running, and walking, and has a wide range of applications [9]. ...
Article
Full-text available
Human Activity Recognition(HAR) plays an important role in the field of ubiquitous computing, which can benefit various human-centric applications such as smart homes, health monitoring, and aging systems. Human Activity Recognition mainly leverages smartphones and wearable devices to collect sensory signals labeled with activity annotations and train machine learning models to recognize individuals’ activity automatically. In order to deploy the Human Activity Recognition model in real-world scenarios, however, there are two major barriers. Firstly, sensor data and activity labels are traditionally collected using special experimental equipment in a controlled environment, which means fitting models trained with these datasets may result in poor generalization to real-life scenarios. Secondly, existing studies focus on single or a few modalities of sensor readings, which neglect useful information and its relations existing in multimodal sensor data. To tackle these issues, we propose a novel activity recognition model for multimodal sensory data fusion: Marfusion, and an experimental data collection platform for HAR tasks in real-world scenarios: MarSense. Specifically, Marfusion extensively uses a convolution structure to extract sensory features for each modality of the smartphone sensor and then fuse the multimodal features using the attention mechanism. MarSense can automatically collect a large amount of smartphone sensor data via smartphones among multiple users in their natural-used conditions and environment. To evaluate our proposed platform and model, we conduct a data collection experiment in real-life among university students and then compare our Marfusion model with several other state-of-the-art models on the collected datasets. Experimental Results do not only indicate that the proposed platform collected Human Activity Recognition data in the real-world scenario successfully, but also verify the advantages of the Marfusion model compared to existing models in Human Activity Recognition.
... Both maternal sensitivity and motherinfant reciprocity are linked to better emotion regulation and attachment security in children (Feldman et al., 2014;Raikes et al., 2007). Parent-infant interactions are a time for infants to receive needed physical contact, including holding, which reduces distress (Karasik et al., 2015;Yao et al., 2019). Finally, the quality of parent-child interactions is predictive of infant brain function across development (Bernier et al., 2016). ...
Article
Much of infant development occurs in the home and in the context of caregiving support. Babies learn through their everyday interactions with parents—from watching, listening, communicating, cuddling, and playing with them. Foundations for cognitive skills such as attention, perception, learning, and language are all built in the brain during the first year of life. Socioemotional development, including the ability to self-regulate behaviors and emotions, also begins during infancy. Recent advances have allowed researchers to answer questions about the developing brain and how it is impacted by experience and environmental systems, including parental sensitivity and consistency, the home environment, socio-cultural factors, community support systems, and public policies. Giving parents the opportunity to support healthy infant development through paid parental leave programs that are accessible, flexible, and equitable, will positively impact early trajectories of brain and behavioral development.
Article
Audio recorders, accelerometers, and cameras that infants wear throughout their everyday lives capture the experiences that are available to shape development. Using sensors to capture behaviors in natural settings can reveal patterns within the everyday hubbub that are unknowable using methods that capture shorter, more isolated, or more planned slices of behavior. Here, we review 10 lessons learned from recent endeavors in which researchers neither designed nor participated in infants’ experiences and instead quantified patterns that arose within infants’ own spontaneously arising everyday experiences. The striking heterogeneity of experiences—the fact that there is no meaningfully “representative” hour of a day, instance of a category, interaction context, or infant—inspires next steps in theory and practice that embrace the complex, dynamic, and multiple pathways of human development.
Article
The study of neonatal cry signals is always an interesting topic and still researcher works interminably to develop some module to predict the actual reason for the baby cry. It is really hard to predict the reason for their cry. The main focus of this paper is to develop a Dense Convolution Neural network (DCNN) to predict the cry. The target cry signal is categorized into five class based on their sound as “Eair”, “Eh”, “Neh”, “Heh” and “Owh”. Prediction of these signals helps in the detection of infant cry reason. The audio and speech features (AS Features) were exacted using Mel-Bark frequency cepstral coefficient from the spectrogram cry signal and fed into DCNN network. The systematic DCNN architecture is modelled with modified activation layer to classify the cry signal. The cry signal is collected in different growth phase of the infants and tested in proposed DCNN architecture. The performance of the system is calculated through parameters accuracy, specificity and sensitivity are calculated. The output of proposed system yielded a balanced accuracy of 92.31%. The highest accuracy level 95.31%, highest specificity level 94.58% and highest sensitivity level 93% attain through proposed technique. From this study, it is concluded that the proposed technique is more efficient in detecting cry signal compared to the existing techniques.
Article
Full-text available
Remote in-home infant monitoring technologies hold great promise for increasing the scalability and safety of infant research (including in regard to the current Covid-19 pandemic), but remain rarely employed. These technologies hold a number of fundamental challenges and ethical concerns that need addressing to aid the success of this fast-growing field. In particular, the responsible development of such technologies requires caregiver input. We conducted a survey of the opinions of 410 caregivers on the viability, privacy and data access of remote in-home monitoring technologies and study designs. Infant-friendly wearable devices (such as sensing body suits) were viewed favourably. Caregivers were marginally more likely to accept video and audio recording in the home if data was anonymised (through automated processing) at point of collection, particularly when observations were lengthy. Caregivers were more open to international data sharing for anonymous data. Caregivers were interested in viewing all types of data, but were particularly keen to access video and audio recordings for censoring purposes (i.e., to delete data segments). Taken together, our results indicate generally positive attitudes to remote in-home monitoring technologies and studies for infant research but highlight specific considerations such as safety, privacy and family practicalities (e.g. multiple caregivers, visitors and varying schedules) that must be taken into account when developing future studies.
Article
Mobile and wearable sensors provide a unique opportunity to capture the daily activities and interactions that shape developmental trajectories, with potential to revolutionize the study of development (de Barbaro, 2019). However, developmental research employing sensors is still in its infancy, and parents’ comfort using these devices is uncertain. This exploratory report assesses parent willingness to participate in sensor studies via a nationally representative survey (N = 210) and live recruitment of a low-income, minority population for an ongoing study (N = 359). The survey allowed us to assess how protocol design influences acceptability, including various options for devices and datastream resolution, conditions of data sharing, and feedback. By contrast, our recruitment data provided insight into parents’ true willingness to participate in a sensor study, with a protocol including 72 h of continuous audio, motion, and physiological data. Our results indicate that parents are relatively conservative when considering participation in sensing studies. However, nearly 41 % of surveyed parents reported that they would be at least somewhat willing to participate in studies with audio or video recordings, 26 % were willing or extremely willing, and 14 % reported being extremely willing. These results roughly paralleled our recruitment results, where 58 % of parents indicated interest, 29 % of parents scheduled to participate, and 10 % ultimately participated. Additionally, 70 % of caregivers stated their reason for not participating in the study was due to barriers unrelated to sensing while about 25 % noted barriers due to either privacy concerns or the physical sensors themselves. Parents’ willingness to collect sensitive datastreams increased if data stayed within the household for individual use only, are shared anonymously with researchers, or if parents receive feedback from devices. Overall, our findings suggest that given the correct circumstances, mobile sensors are a feasible and promising tool for characterizing children’s daily interactions and their role in development.
Article
Research in sensor based human activity recognition (HAR) has been a core concern of the mobile and ubiquitous computing community. Sophisticated systems have been developed with the main view on applications of HAR methods in research settings. This work addresses a related yet practically different problem that mainly focuses on users of HAR technology. We acknowledge that practitioners from outside the core HAR research community are motivated to employ HAR methods for practical deployments. Even though standard processing approaches exist, arguably, often times substantial modifications are necessary to derive effective analysis systems. It is not always clear a-priori how challenging a HAR task actually is and what dimensions of an analysis pipeline are crucial for successful automated assessments. In practice this can lead to disappointing results or disproportionate efforts that have to be invested into the optimization of data analysis pipelines, that were supposed to work "out of the box". We present a framework for the objective complexity assessment of HAR tasks that directly supports practitioners' decision making of whether and how to employ HAR for their deployments. We map a HAR task onto a vectorial representation that allows us to analyse the inherent challenges of the task and to draw conclusions through similarity analysis with regards to existing tasks. We validate our complexity assessment framework on 23 HAR datasets and derive a data-driven categorization of human activity recognition. We demonstrate how our objective analysis can be used to inform the deployment of HAR systems in practical scenarios.
Article
Full-text available
Responsive feeding—initiating feeding in response to early hunger cues—supports the physiology of lactation and the development of infant feeding abilities, yet there is a dearth of research examining what predicts responsive feeding. In non-Western proximal care cultures, there is an association between responsive feeding and mother–infant physical contact, but this has not been investigated within Western populations. In two studies, we tested whether mother–infant physical contact predicted feeding in response to early hunger cues versus feeding on a schedule or after signs of distress among U.S. breastfeeding mothers. With an online questionnaire in Study 1 (n = 626), physical contact with infants (via co-sleeping and babywearing) predicted increased likelihood of self-reported responsive feeding. Mothers who reported responsive feeding were more likely to exclusively breastfeed for the first six months, breastfeed more frequently throughout the day, and had a longer planned breastfeeding duration than mothers who reported feeding on a schedule or after signs of infant distress. In Study 2 (n = 96), a three-day feeding log showed that mother–infant physical contact predicted feeding in response to early hunger cues but mother–infant proximity (without physical contact) did not. In sum, our results demonstrate that physical contact with infants may shape breastfeeding behavior among U.S. mothers, highlighting a connection between social interaction and infant nutrition that warrants further investigation.
Article
Full-text available
Objectives: Stressful family environments early in life have negative effects on physical health. However, less is known about the health effects of positive aspects of families. We examined the associations between maternal responsiveness and immune markers among youth with asthma and identified youth expressions of positive affect as a potential mechanism of these associations. Methods: Forty-three youths with asthma (26 boys; aged 10-17 years) wore the Electronically Activated Recorder for 4 days to assess maternal responsiveness and youth expressions of affect from audio-recordings of daily life. Trained coders rated Electronically Activated Recorder sound files for expressions of maternal responsiveness and affect displayed by the youth. Peripheral blood mononuclear cells were isolated, cultured, and assayed to determine stimulated levels of interleukin (IL)-5, IL-13, and interferon-γ. Results: Greater maternal responsiveness was associated with decreased stimulated production of IL-5 (r = -0.38, p = .012) and IL-13 (r = -0.33, p = .031). Greater total positive affect in youth was linked to decreased stimulated production of IL-5 (r = -0.46, p = .002) and IL-13 (r = -0.37, p = .014). Total negative affect among youth was unrelated to immune responses. There was a significant indirect effect of maternal responsiveness via positive affect in youth on lower levels of IL-5 (95% confidence interval = -3.41 to -0.03) and IL-13 (95% confidence interval = -2.34 to -0.01) when adjusting for caregiver-youth conflict and negative affect among youth. Conclusions: These results indicate the importance of positive family interactions for youth and provide preliminary evidence for a mechanism through which parenting can influence immune responses in youth with asthma.
Conference Paper
Full-text available
Management of Parkinson's Disease (PD) could be improved significantly if reliable, objective information about fluctuations in disease severity can be obtained in ecologically valid surroundings such as the private home. Although automatic assessment in PD has been studied extensively, so far no approach has been devised that is useful for clinical practice. Analysis approaches common for the field lack the capability of exploiting data from realistic environments, which represents a major barrier towards practical assessment systems. The very unreliable and infrequent labelling of ambiguous, low resolution movement data collected in such environments represents a very challenging analysis setting, where advances would have significant societal impact in our ageing population. In this work we propose an assessment system that abides practical usability constraints and applies deep learning to differentiate disease state in data collected in naturalistic settings. Based on a large data-set collected from 34 people with PD we illustrate that deep learning outperforms other approaches in generalisation performance, despite the unreliable labelling characteristic for this problem setting, and how such systems could improve current clinical practice.
Article
Rapidly maturing technologies for sensing and activity recognition can provide unprecedented access to the complex structure daily activity and interaction, promising new insight into the mechanisms by which experience shapes developmental outcomes. Motion data, autonomic activity, and “snippets” of audio and video recordings can be conveniently logged by wearable sensors (Lazer et al., 2009). Machine learning algorithms can process these signals into meaningful markers, from child and parent behavior to outcomes such as depression or teenage drinking. Theoretically motivated aspects of daily activity can be combined and synchronized to examine reciprocal effects between children’s behaviors and their environments or internal processes. Captured over longitudinal time, such data provide a new opportunity to study the processes by which individual differences emerge and stabilize. This paper introduces the reader to developments in sensing and activity recognition with implications for developmental phenomena across the lifespan, sketching a framework for leveraging mobile sensors for transactional analyses that bridge micro‐ and longitudinal‐ timescales of development. It finishes by detailing resources and best practices to facilitate the next generation of developmentalists to contribute to this emerging area.
Article
Baby wearable technologies are becoming increasingly popular, particularly in early infancy. However, little research has been conducted to investigate how baby wearable technologies are adopted and used in parenting. This paper presents a two-week in-depth situated study with six mothers in their homes consisting of contextual entry and exit interviews, video recordings of "out-of-box" experiences, and a diary study. Using interpretative phenomenological analysis, participants' use and expectations of the baby wearable technology were examined. Use of the device directly impacted upon parents' knowledge production and anxiety, and influenced the perceived physicality and social factors of parenting. We frame these findings around sociological norms of the vulnerable child and responsible mother, as well as the notion of "lived informatics", where individuals' interaction with the technology influenced the perception, use and impact of the baby wearable on everyday parenting practices and familial relationships.
Article
Background: Kangaroo mother care (KMC), originally defined as skin-to-skin contact between a mother and her newborn, frequent and exclusive or nearly exclusive breastfeeding, and early discharge from hospital, has been proposed as an alternative to conventional neonatal care for low birthweight (LBW) infants. Objectives: To determine whether evidence is available to support the use of KMC in LBW infants as an alternative to conventional neonatal care before or after the initial period of stabilization with conventional care, and to assess beneficial and adverse effects. Search methods: We used the standard search strategy of the Cochrane Neonatal Review Group. This included searches in CENTRAL (Cochrane Central Register of Controlled Trials; 2016, Issue 6), MEDLINE, Embase, CINAHL (Cumulative Index to Nursing and Allied Health Literature), LILACS (Latin American and Caribbean Health Science Information database), and POPLINE (Population Information Online) databases (all from inception to June 30, 2016), as well as the WHO (World Health Organization) Trial Registration Data Set (up to June 30, 2016). In addition, we searched the web page of the Kangaroo Foundation, conference and symposia proceedings on KMC, and Google Scholar. Selection criteria: Randomized controlled trials comparing KMC versus conventional neonatal care, or early-onset KMC versus late-onset KMC, in LBW infants. Data collection and analysis: Data collection and analysis were performed according to the methods of the Cochrane Neonatal Review Group. Main results: Twenty-one studies, including 3042 infants, fulfilled inclusion criteria. Nineteen studies evaluated KMC in LBW infants after stabilization, one evaluated KMC in LBW infants before stabilization, and one compared early-onset KMC with late-onset KMC in relatively stable LBW infants. Sixteen studies evaluated intermittent KMC, and five evaluated continuous KMC. KMC versus conventional neonatal care: At discharge or 40 to 41 weeks' postmenstrual age, KMC was associated with a statistically significant reduction in the risk of mortality (risk ratio [RR] 0.60, 95% confidence interval [CI] 0.39 to 0.92; eight trials, 1736 infants), nosocomial infection/sepsis (RR 0.35, 95% CI 0.22 to 0.54; five trials, 1239 infants), and hypothermia (RR 0.28, 95% CI 0.16 to 0.49; nine trials, 989 infants; moderate-quality evidence). At latest follow-up, KMC was associated with a significantly decreased risk of mortality (RR 0.67, 95% CI 0.48 to 0.95; 12 trials, 2293 infants; moderate-quality evidence) and severe infection/sepsis (RR 0.50, 95% CI 0.36 to 0.69; eight trials, 1463 infants; moderate-quality evidence). Moreover, KMC was found to increase weight gain (mean difference [MD] 4.1 g/d, 95% CI 2.3 to 5.9; 11 trials, 1198 infants; moderate-quality evidence), length gain (MD 0.21 cm/week, 95% CI 0.03 to 0.38; three trials, 377 infants) and head circumference gain (MD 0.14 cm/week, 95% CI 0.06 to 0.22; four trials, 495 infants) at latest follow-up, exclusive breastfeeding at discharge or 40 to 41 weeks' postmenstrual age (RR 1.16, 95% CI 1.07 to 1.25; six studies, 1453 mothers) and at one to three months' follow-up (RR 1.20, 95% CI 1.01 to 1.43; five studies, 600 mothers), any (exclusive or partial) breastfeeding at discharge or at 40 to 41 weeks' postmenstrual age (RR 1.20, 95% CI 1.07 to 1.34; 10 studies, 1696 mothers; moderate-quality evidence) and at one to three months' follow-up (RR 1.17, 95% CI 1.05 to 1.31; nine studies, 1394 mothers; low-quality evidence), and some measures of mother-infant attachment and home environment. No statistically significant differences were found between KMC infants and controls in Griffith quotients for psychomotor development at 12 months' corrected age (low-quality evidence). Sensitivity analysis suggested that inclusion of studies with high risk of bias did not affect the general direction of findings nor the size of the treatment effect for main outcomes. Early-onset KMC versus late-onset KMC in relatively stable infants: One trial compared early-onset continuous KMC (within 24 hours post birth) versus late-onset continuous KMC (after 24 hours post birth) in 73 relatively stable LBW infants. Investigators reported no significant differences between the two study groups in mortality, morbidity, severe infection, hypothermia, breastfeeding, and nutritional indicators. Early-onset KMC was associated with a statistically significant reduction in length of hospital stay (MD 0.9 days, 95% CI 0.6 to 1.2). Authors' conclusions: Evidence from this updated review supports the use of KMC in LBW infants as an alternative to conventional neonatal care, mainly in resource-limited settings. Further information is required concerning the effectiveness and safety of early-onset continuous KMC in unstabilized or relatively stabilized LBW infants, as well as long-term neurodevelopmental outcomes and costs of care.
Article
In this article, we describe the design process and principles used in the development of Estrellita, a tool to support parents of preterm infants to track health data. We tested Estrellita in the homes of seven families for 4 months while following seven additional families without Estrellita. The feedback from this trial, including in-depth interviews, surveys, and log analyses, sheds light on how parents can use a mobile data collection tool to enhance their problem-solving processes about their own health and that of their infants, as well as to share with others who support them in this care. In addition to presenting the design of a recording technology for preterm infants and its use in a real-life setting, the results of this research provide a deep understanding of how technology can and should be used to support home care of at-risk patients, in which data capture may be essential.
Conference Paper
In this paper we demonstrate a Markov model based technique for recognizing gestures from accelerometers that explicitly represents duration. We do this by embedding an Erlang-Cox state transition model, which has been shown to accurately represent the first three moments of a general distribution, within a Dynamic Bayesian Network (DBN). The transition probabilities in the DBN can be learned via Expectation-Maximization or by using closed-form solutions. We test this modeling technique on 10 hours of data collected from accelerometers worn by babies pre-categorized as high-risk in the Newborn Intensive Care Unit (NICU) at UCI. We show that by treating instantaneous machine learning classification values as observations and explicitly modeling duration, we improve the recognition of Cramped Synchronized General Movements, a motion highly correlated with an eventual diagnosis of Cerebral Palsy.
Article
This paper presents a child activity recognition approach using a single 3-axis accelerometer and a barometric pressure sensor worn on a waist of the body to prevent child accidents such as unintentional injuries at home. Labeled accelerometer data are collected from children of both sexes up to the age of 16 to 29 months. To recognize daily activities, mean, standard deviation, and slope of time-domain features are calculated over sliding windows. In addition, the FFT analysis is adopted to extract frequency-domain features of the aggregated data, and then energy and correlation of acceleration data are calculated. Child activities are classified into 11 daily activities which are wiggling, rolling, standing still, standing up, sitting down, walking, toddling, crawling, climbing up, climbing down, and stopping. The overall accuracy of activity recognition was 98.43% using only a single- wearable triaxial accelerometer sensor and a barometric pressure sensor with a support vector machine.