Mid Sweden University
A Survey of Wireless Sensor Networks for Home Healthcare Monitoring
Application
Jun Tang and Tingting Zhang
Department of Information Technology and Media, Mid Sweden University, Sundsvall, Sweden
Keywords: WSN, Healthcare Monitoring Application, Home Care.
Abstract: In recent years, wireless sensor network technology has become mature. Working together with biomedical
engineering, it has enormous potential benefits to improve the lifestyle of human especially for the elderly.
This survey mainly focuses on two prototypes of home healthcare monitoring application: daily activities
monitoring application and medical status monitoring application. It will present the requirement analysis
which starts from the causes of chronic diseases. The paper also discusses challenges for current home
healthcare monitoring application. At the last part of the survey, it will give the conclusion and future
aspects.
1 INTRODUCTION
In recent decades, the elderly population increases
continuously. There will be nearly 120 million
people at the age 65 or more than 65 years old in
2050 (Kinsella and Philips, 2005). As a consequence
of the population aging, the old people will has more
possibility to suffer the difficulties from physical
disability, health problem and cognitive impairment
(Lee et al., 2010), which not only make the old
people cannot live independently but also give them
a huge economic burden. So, finding a good way to
cure the chronic diseases and reduce the economic
burden of the old people is an important issue.
However, working together with biomedical
engineering, wireless sensor network technology has
enormous potential benefits to improve the lifestyle
of human especially for the elderly. The design
consideration of WSN healthcare system provides
new possibility to response to population aging.
This survey starts from the causes of chronic
diseases and focus on the detail requirement in two
prototypes of homecare monitoring application:
daily activities monitoring application and medical
status monitoring application. The paper also put
forward the challenges for current healthcare
monitoring application. At the last part of the
survey, it will give the conclusion and future
aspects.
2 PROTOTYPES OF HOME
HEALTHCARE MONITORING
APPLICATIONS
There are not many prototypes for healthcare
monitoring application. According to the purposes of
healthcare monitoring application, prototype can be
generally divided into two categories: medical status
monitoring applications and daily activities
monitoring applications. The former is that using
biomedical sensors such as blood pressure sensor,
ECG sensor and heart rate sensor to monitor the
physiological parameters of subjects in order to get
users’ health status. While the later one using mostly
the environmental sensors to monitor the activities
of daily living activities for the subjects. The
following subsections will give examples of these
two prototypes of home care monitoring system.
2.1 Medical Status Monitoring
Applications
The home care monitoring application area, most of
studies focus on medical status monitoring which
collects the vital signs of subject in order to allow
user to understand their health status and make
caregivers easy to manage the medical information.
CodeBlue (Lab, 2008) is developed by Harvard
Sensor Networks Lab which is a wireless sensor for
240
Tang J. and Zhang T..
Mid Sweden University - A Survey of Wireless Sensor Networks for Home Healthcare Monitoring Application.
DOI: 10.5220/0004265702400243
In Proceedings of the 2nd International Conference on Sensor Networks (SENSORNETS-2013), pages 240-243
ISBN: 978-989-8565-45-7
Copyright
c
2013 SCITEPRESS (Science and Technology Publications, Lda.)
medical care application including both hardware
platforms and software platforms. Wireless pulse
oximeter sensor and wireless two-lead ECG sensor
is used based on the TinyOS operating system, so
that it is supported to monitor a variety of
physiological parameters include heart rate (HR),
oxygen saturation (SpO2), and ECG data. The user
can get their health status without going to the
hospital. The software platforms support PDAs, PCs
and other devices which make the user and caregiver
to manage these medical data easily. CodeBlue
system also supports indoor and outdoor location
tracking, so that it can be used to identify user’s
location especially helpful for the people with
cognitive disabilities.
With the development of biomedical sensor,
there are a lot of physiological variables can be
monitored. For example, in (Yang et al., 2010)
(Morris et al., 2009), by using biochemical sensing
techniques, a textile-based wearable biosensor can
monitor pH and sodium (Na+) when the user wear it.
2.2 Daily Activities Monitoring
Applications
Activities monitoring applications at first are mainly
targeted for gerontology. The Geriatrics shows that
with the increase of age, elderly body systems, such
as nervous system and immune system will
gradually decline (Wikipedia, 2012). The simple and
inexpensive environment sensors are used to
monitor the daily life of elderly. Such applications
require the long-term monitoring, collecting and
analyzing data to find out when a person begins to
lose physical or mental ability, or discover the trends
of chronic disease, so that it have the potential to
prevent the elderly physical and mental illness.
There is a study in the Quality of Life
Technology Center (QoLTC) of Carnegie Mellon
University that using embedded sensors to
monitoring observations of daily living (ODLs) to
detect onset of Alzheimer's disease (Mellon, 2010).
They proposed a new concept Observations of Daily
Living, or ODLs which includes "patient-recorded
feelings, thoughts, behaviours and environmental
factors (Mellon, 2010)". They proposed a new
concept Observations of daily living, or ODLs
which include "patient-recorded feelings, thoughts,
behaviours and environmental factors (Mellon,
2010)". And they put environment wireless sensors
in about 50 old adults’ house to monitor their daily
activities in order to detect subtle changes in
behaviour.
3 REQUIREMENT ANALYSIS
FOR HOME HEALTHCARE
MONITORING APPLICATION
Most of the healthcare system is designed for the
elderly and the people who suffer from chronic
disease. But when some researchers design the
healthcare system, there is a problem that they did
not give the motivations or reasons the physiological
parameters or physical activities to be monitored.
For example, the daily activity such as teeth-
brushing or shaving (Ince et al., 2008) may not
provide the useful information for the doctor. If the
collected data have no effect to detect or cure the
chronic disease, or if the collected data have less
value to find out the activity patterns, there is no
need to monitor these parameters. Therefore, this
section provides requirement analysis which starts
from the causes of chronic diseases and focus on
what activities and physiological parameters which
have more medical value for several chronic
diseases.
“Table 1” shows the summary of all the user
requirements analysis about the activities and
physiological parameters to be monitored for
chronically ill mentioned here. Researchers can use
this table as a reference when they design health care
system both for the medical status monitoring
applications and the daily activities monitoring
applications.
3.1 Diabetes
For the people who have diabetes, they always have
high blood glucose level and produce the classical
symptoms of polyuria (frequent urination),
polydipsia (increased thirst) and polyphagia
(increased hunger). Beside of these three symptoms,
there are other sign of diabetes: constant hunger,
unexplained weight loss, flu-like symptoms,
including weakness and fatigue, blurred vision and
slow healing of cuts or bruises. From WHO diabetes
criteria, if the one whose fasting glucose is over 125
mg/dl and 2 hour glucose is at or above 200 mg/dl,
he or she can be diagnosed with diabetes.
3.2 Cardiovascular Disease
Cardiovascular disease is related with heart or blood
vascular disease. The two main causes are high
blood pressure and high blood cholesterol. Other
important risk factors are diabetes, tobacco use, poor
nutrition, and overweight and obesity. There are
MidSwedenUniversity-ASurveyofWirelessSensorNetworksforHomeHealthcareMonitoringApplication
241
three means to diagnosis cardiovascular disease that
are blood tests, electrophysiology and medical
imaging. However, blood tests are minimally
invasive and medical imaging is expensive and only
work in hospital, so the better approach is to
measure the Electrocardiogram (ECG). At the same
time, blood pressure and heart rate monitoring are
also necessary.
3.3 Epilepsy
Epilepsy is a chronic neurological disease which can
occur anytime so that the detection of seizures is
important. Figure 1 shows an electroencephalogram
(EEG) that contains EEG signal and Epileptic
seizure EEG signal. This figure tells us that it can
use EEG to monitor seizures.
Figure 1: Normal EEG signal and Epileptic seizure EEG
signal (Lange, 2005).
3.4 Alzheimer's Disease
Alzheimer's disease (AD), also called senile
dementia, has symptoms such as confusion,
irritability and aggression, mood swings, language
breakdown; long-term memory loss and senses
decline. At present, the early detection of
Alzheimer's disease is monitoring the movement
pattern.
3.5 Obesity
Obesity is caused by excessive intake of energy
from food but not getting enough physical activity.
As the consequence of obesity, it may lead to
diabetes, heart disease high blood pressure and sleep
and respiratory problems. Bad habits may lead to
chronic disease. So it is important to monitor the
daily behaviours which for obesity are diet, physical
activity, alcohol and tobacco use and medicine
intaking.
Table 1: Requirements analysis about the activities and
physiological parameters to be monitored for chronically
ill.
Disease
Activities to be
monitored
Physiological
p
arameters to be
monitored
Diabetes
Quantity of water
intaking
Blood glucose
Quantity and
frequency of eating
Blood pressure
Quantity of smoking Cholesterol
Frequency of
urination
Body weight
Medicine intaking Vision
Cardiovascular
Disease
Alcohol
consumption
ECG
Tobacco cessation
Blood
cholesterol
Daily activity of
moderate to vigorous
exercise
Blood pressure
Emotional stress in
day to day life
Body weight
Epilepsy
Bradykinesia EEG
Idiosyncratic
motions
ECG
Alzheimer's disease Quantity of smoking
Blood glucose
Blood pressure
Cholesterol
4 CHALLENGES OF THE
CURRENT HOME
HEALTHCARE MONITORING
APPLICATION
When we get the benefit from the wireless sensor
network healthcare applications, we also face many
challenges. This section presents the challenges and
problems for current home healthcare monitoring
applications. In this survey, the challenges and
current problems are discussed from both the user
and developer sides.
4.1 User Level Challenges
For the elderly who suffer from chronic disease, they
prefer non-invasive measurement sensor rather than
minimally invasive measurement sensor. For
example, the most popular way to measure blood
glucose is pricking the finger and getting some
blood, then using blood glucose monitors to get the
blood glucose level. But most of the users don’t
want this minimally invasive measurement. So, to
SENSORNETS2013-2ndInternationalConferenceonSensorNetworks
242
solve this problem, developer needs to find a better
way of measuring.
For the caregiver, the health care application
should be easy-to-learn and easy-to-operate. This
problem becomes one of the reasons of the failure of
Google health. The healthcare data type of Google
health is too complex to use, so that users are
deterred by the burden of data entry. So make the
healthcare application easy-to-use is very important.
For healthcare professional, the different areas of
experts may have different requirements. Vascular
doctor has said that have the requirement of blood
flow sensor. Because cells or tissues necrosis caused
by inadequate blood flow. So his patients have to
amputees at this situation. Therefore, there is still
much developing space for WSN healthcare system.
Developer had better to find out the current needs
before they start the research.
4.2 Developer Level Challenges
The most important issue at the developer lever is
that the biomedical sensors are directly in contact
with the human body, which is different from any
other common sensors. However, most of the WSN
healthcare systems is still under research and
develop in laboratories (Ren and Meng, 2006). So,
they cannot be directly applied to the healthcare
system because of the biological effects that may
arise. So, check out whether the biological effects
have any effect for human body is an important task
for developer.
The healthcare application is better to have high
interoperability. For example, different kind of
sensor may use different communication protocol
which is hard to work together in one system.
Sometimes, an application may design its own
communication protocol lead to the problem that
researcher development platform cannot unify.
Therefore, it is necessary to develop middleware to
improve system interoperability.
5 CONCLUSIONS AND FUTURE
ASPECTS
The paper presented a detail requirement analysis
starting from the causes of chronic diseases and
made a summary about five typical chronic diseases
and analyzed the user requirement and gave results
at “Table 1” that show the user requirement about
the activities and physiological parameters to be
monitored for chronic disease. This survey discussed
the challenges from both user and developer side. In
the future, finding a good way to solve these
problems will make the WSN healthcare system
works better to improve people's lifestyles.
REFERENCES
D. Morris, S. Coyle, Y. Wu, K. T. Lau, G. Wallace, and D.
Diamond. 2009. Bio-sensing textile based patch with
integrated optical detection system for sweat
monitoring. In Sen. Act. B: Chem. vol. 139, 2009, Vol.
no. 1, pp. 231–236.
Hongliang Ren, Max Q.-H. Meng. 2006. Bioeffects
Control in Wireless Biomedical Sensor Networks. In
Biomedical Enginering Lab, Department of Electronic
Engineering. s.l. : IEEE, 2006.
Kevin Kinsella, David R. Phillips. 2005. Global Aging:
The Challenge of Success. u.o. In Population
Reference Bureau, 2005.
Lab, Harvard Sensor Networks. 2008. CodeBlue: Wireless
Sensors for Medical Care. [Online] 2008.
http://fiji.eecs.harvard.edu/CodeBlue.
Lange, Der. 2005. File:Spike-waves.png. [Online] den 11
6 2005. http://en.wikipedia.org/wiki/File:Spike-
waves.png.
Mellon, Carnegie. 2010. Carnegie Mellon to Test Ability
of Embedded Sensors to Detect Onset of Dementia,
Infirmity. [Online]
http://www.hcii.cmu.edu/news/stories/carnegie-
mellon-test-ability-embedded-sensors-detect-onset-
dementia-infirmity.
N. F. Ince, C. H. Min, A. Tewfik, D. Vanderpool. 2008.
Detection of early morning daily activities with static
home and wearable wireless sensors. In EURASIP
Journal on Advances in Signal Processing. 2008.
Ronald Lee, Andrew Mason and Daniel Cotlear. 2010.
SOME ECONOMIC CONSEQUENCES OF
GLOBAL AGING. In Network : e Health, Nutrition,
and Population Family (HNP), 2010.
Wikipedia. 2012. Gerontology.
http://en.wikipedia.org/wiki/Gerontology. [Online] den
7 11 2012. [Citat: den 11 11 2012.]
Y. L. Yang, M. C. Chuang, S. L. Lou, and J. Wang. 2010.
Thick-film textile-based amperometric sensors and
biosensors. In Analyst. vol. 135, 2010, Vol. pp. 1230–
1234.
MidSwedenUniversity-ASurveyofWirelessSensorNetworksforHomeHealthcareMonitoringApplication
243
