A Novel Mobile-Cloud based Healthcare Framework for Diabetes
Maria Salama
1
and Ahmed Shawish
2
1
Department of Computer Science, British University in Egypt, Cairo, Egypt
2
Department of Scientific Computing, Ain Shams University, Cairo, Egypt
Keywords: e-Health, Healthcare, Diabetes, Cloud Computing, Mobile Cloud Computing.
Abstract: Healthcare is one of the most important sectors in all countries; consuming an average of 9.5% of their
domestic product. Among the widely spread endemic diseases, Diabetes is particularly a non-cured one that
consumes medical resources, follow-up efforts, regular set of diverse checkups, continual needs for
physicians and medical supplies. The current advanced healthcare systems succeeded to be a reservoir of
healthcare records focussing on emergency and medical imaging only. At the same time, most of the
currently deployed systems cannot provide a daily follow up to patients as well as not being able to help
governments to smartly allocate their medical resources. In this paper, we propose a comprehensive
framework that incorporates Mobile and Cloud computing with data mining techniques to efficiently
provide a real-time smart healthcare framework for Diabetes. Mobile application is emerged as a widely
deployed communication tool between the patients and the proposed system hosted on the Cloud. The
Cloud is incorporated to accommodate the system due to its known features that makes it possible to acquire
continually fresh data from the field. The data is then processed through smart data mining techniques to
extract knowledge and draw conclusions for governments. As discussed in the paper, the proposed
framework promises a significant enhancement in the resources allocation and utilization, as well as
provides a faster emergency response. The proposed architecture has been simulated using Junosphere, a
cloud simulator, and it could represent a significant step towards future smart healthcare systems specialized
to cover other endemic diseases.
1 INTRODUCTION
Countries around the globe spends considerable
amount of its resources to provide the best
healthcare services to its citizens. Based on 2012
statistics mentioned in (Organization of economics
co-operation and development, 2013), developed
countries consumed an average of 9.5% of their
gross domestic product. For instance, The United
States (17.6%), Netherlands (12%), France and
Germany (11.6%) were the top four spenders in this
sector. One of the most critical diseases that daily
consume lot of resources is the Diabetes; as one of
the main widespread non-cured diseases. According
to the New York Time’s magazine, developing
countries have a huge percentage of Diabetics that
reaches almost 42% of the population. Many of them
experience early-stage eye disease, while about 5%
are totally blind (Hamdan, 2011). Such high
percentages are mainly due to the bad management
of the diseases. The only way to avoid impacts of the
diabetes is to keep it under regular supervision.
Diabetics suffer from their inability to easily
manage their treatments; as it regularly involves
taking medications, watching their blood glucose
level, keeping their records as well as regularly
visiting their physician and performing medical
checkups. These boring processes and long queues
in medical facilities affect the patients’ life-style and
by time most of the patients gradually start to
neglect their treatments and surround to the
implications. Despite the usage of sophisticated
healthcare systems in many countries (Löhr,
Sadeghi and Winandy, 2010) (Wooten, Klink, Sinek,
Bai, & Shar, 2012), this disease is still handled as
any disease without efficiently controling this drain
of resources. Traditional data collection methods are
still used to inspect the diabetes development in the
field. Such methods result by their turn in slow
feedback cycle and hence obsolete conclusions as
well as inefficient action plans. Even, most of the
available computer-based solutions can be described
as a huge repository of data without real analysis or
262
Salama M. and Shawish A..
A Novel Mobile-Cloud based Healthcare Framework for Diabetes.
DOI: 10.5220/0004750702620269
In Proceedings of the International Conference on Health Informatics (HEALTHINF-2014), pages 262-269
ISBN: 978-989-758-010-9
Copyright
c
2014 SCITEPRESS (Science and Technology Publications, Lda.)
knowledge discovery as it should be.
The need of a smart system that adopts recent IT-
technologies with up-to-date communication
technologies becomes crucial to cope with this
disease and efficiently utilize the available medical
resources to provide a better healthcare service to
the increasing number of diabetes patients.
In this paper, we propose a comprehensive
framework that incorporates Mobile Cloud
computing with data mining techniques to efficiently
provide a smart healthcare system for diabetes. The
smart mobile, as an efficient channel with the
patients, is used to vastly acquire patients data at low
expenses. The Diabetes mobile application will help
patients to fully mange their daily treatment process
and will deliver guiding instructions. The Cloud is
incorporated to accommodate the system, due to its
known features, that makes it possible to acquire
real-time diabetes patients’ data to be then processed
through smart data mining techniques for knowledge
discovery. The integrated GPS service available on
smart mobiles has been incorporated in the
framework to provide patient’s spatial location for
fast emergency response. As illustrated and
discussed through case study, the proposed
framework promises a significant enhancement in
the health resources allocation and utilization and
demonstrates advantages offered to diabetes.
The rest of this paper is organized as follows.
Section 2 briefs about Diabetes and its management.
Section 3 introduces the background on the mobile-
based management systems and current cloud-based
healthcare systems. In section 4, we address the
proposed framework. In section 5, we illustrate the
implementation and simulation of an end-to-end
scenario. Finally, the paper is concluded and our
future work is reflected in section 6.
2 DIABETES MANAGEMENT
In this section, we present a brief about the Diabetes
and the disease management.
Diabetes is considered as one of the main non-
cured diseases, once infected it remains forever.
Diabetes occurs because the body can't use glucose
properly, either owing to a lack of the hormone
insulin or the insulin available is not effective. This
decease has three major types; type I, type II and
Gestational diabetes. Having diabetes increases the
risk of other health problems, but there are lots of
things to do to minimize them; such as maintaining a
healthy lifestyle, attending regular check-ups and
monitoring glucose levels (BBC - Health: Diabetes,
2013). According to the charity Diabetes UK, more
than two million people in UK have the condition
and up to 750,000 more are believed to have it
without realizing (Hamdan, 2011).
In fact, diabetes is a condition that needs to be
managed daily. Diabetes management can refer to
dealing with short term events; such as high and low
blood glucose to control it over the long term by
getting to grips with understanding the condition.
Keeping an eye on the blood glucose levels is the
key to everyday management of type I and II; as it
helps to plan eating, schedule exercise and take
medication (Diabetes matters, 2013). Management
of diabetes involves more than keeping blood
glucose levels under control. It also encompasses
keeping blood pressure and cholesterol levels under
control, maintaining weight and dealing with the
emotional impact of the condition. Diabetes is far
from an easy condition to keep under control to the
extent that there are a number of different
educational courses designed to help people cope
with the daily challenges (Diabetes.co.uk, 2013).
3 BACKGROUND
In this section, we present the diabetes mobile-based
management systems available and the currently
implemented cloud-based e-health systems.
3.1 Diabetes Mobile-based
Management Systems
With the wide use of mobile phones and applications
development in different mobile operating systems,
different applications have been developed in this
field. Known ones are Glucose Buddy, BGluMon,
WaveSense and Vree.
Glucose Buddy is a mobile application in Apple
Store; featuring log for blood glucose, medication,
food and exercise log. Prints out reports are
generated in the form of grid table, without analysis
(glucose buddy, 2011).
BGluMon (Blood Glucose Monitor), found in
Apple Store, is used to watch the patient’s blood
glucose level on daily basis. It includes the functions
of recording and exporting data. Graphs are
generated in an advanced way that cannot be useful
for novice users (bglumon, 2012).
WaveSense developers had a new idea for
measuring blood glucose level; by an invented meter
that can be connected to iPhone or any other Apple
device and can measure blood glucose level through
strips specialized for this meter. After measuring the
ANovelMobile-CloudbasedHealthcareFrameworkforDiabetes
263
blood glucose level, data is saved into WaveSense
application automatically (IBG Star, 2012). But this
meter and its strips are available only in certain
countries; which limit the use of this application.
Vree, also developed for Apple iOS, is
specialized for type II diabetes. This application has
multiple functions; such as glucose, medication and
nutrition tracking; focusing only on type II that none
of its functions and calculations can be used for
other types of diabetes (Vree, 2012).
On the other hand, the only application in the
Android Market is used to get patients information,
manage diabetes by tracking food, medication,
weight and level of glucose in the blood. By tracking
these values, the application can display a summary
of the patient’s progress and hence archive it (On
Track, 2012). Yet, it is limited in other
functionalities such as connectivity.
3.2 Healthcare Cloud-based Systems
As for e-health systems currently implemented,
examples include @HealthCloud, Health Cloud
eXchange (HCX), Emergency Medical System
(EMS), and HealthATM kiosks.
@HealthCloud is a mobile healthcare
information management system that is based on
cloud computing and Android OS. It enables
healthcare data storage, update and retrieval using
Amazon Simple Storage Service (S3). It includes a
PHR application that acquires and displays patient
records stored in the cloud and a medical imaging
module to display medical images on the device. It
also supports native multi-touch technology which
allows better manipulation of medical images
(Doukas, Pliakas and Maglogiannis, 2010).
HCX is a distributed web interactive system that
provides a private cloud-based data sharing service
allowing dynamic discovery of various health
records and related healthcare services. HCX allows
sharing health records between different EHR
systems and. automatically adapts to changes in the
cloud (Mohammed and Fiaidhi, 2010).
EMS is an emergency medical system that
accesses PHRs of patients and helps provide timely
care. It mainly consists of PHRs platform, EMS
application and a Portal to access the former. EMS
uses a private cloud to store data; in particular PHR
data. It helps facilitate a timely access of relevant
information by authorized people in case of
emergencies (Koufi, Malamateniou and
Vassilacopoulos, 2010).
HealthATM kiosks are developed for patients to
manage their own personal health data, integrating
services from Google’s cloud environment. It
provides timely access to relevant health data to
patients and strengthens patients’ communication
with their care providers (Botts, Thoms, Noamani
and Horan, 2010). Although, it is also a cost
effective solution of personal healthcare
management; as they makes use of cloud computing
architectures, the systems currently cannot be
directly handed over to patients; for constant
training, outreach and education are must.
3.3 Discussion
The currently available mobile-based management
systems are limited for personal use only with
scattered capabilities that are not all gathered in one
application. In general, they are designed to help a
single patient to manage his own diabetic case
without any interaction with their physicians.
On the other hand, the cloud-based medical
systems are designed to maximize the benefits of
treatment and emergency intervention enhancement,
while others to save and share healthcare records
especially big sized data like medical images. Some
of them leverage the benefit of combining the
mobile and Cloud. However, this benefit has not
been yet interactively addressed through an
integrated framework that provides a practical
solution for a real medical problem as we do in this
paper.
4 PROPOSED FRAMEWORK
In this section, we present a comprehensive
framework that incorporates Mobile Computing and
Cloud computing with data mining techniques to
efficiently provide a smart healthcare system for
fully supporting Diabetes.
With the fast and widely available smart phones,
the mobile application is meant to be a fast and
portable efficient connection method between the
system and the citizens. Through such
communication channel, data can be vastly and
vigorously acquired from the field at low expenses
and in a parallel manner along the country saving
valuable time and resources. Guiding instructions
are also delivered to the citizens anywhere and
anytime through the same channel. In addition,
diabetes medical sensors are integrated with the
mobile to help the patient in his daily life.
The Cloud, on the other hand, is incorporated to
accommodate the system due to its broadly 24/7
availability, scalability, and huge storage capabilities
HEALTHINF2014-InternationalConferenceonHealthInformatics
264
that makes it possible to support huge number of
concurrent users access on the run. This feature by
its turn makes it possible to acquire continually fresh
updated medical record and feedbacks from citizens.
The acquired data will be then processed through
data mining techniques and knowledge discovery
algorithms to smartly benefit from such huge pool of
fresh data. Important and useful relation between
data can be efficiently deduced; specifically the
spatial distribution of the data sources (known by the
smart mobiles GPS) will bring a long list of
advantages and edge of knowledge in this field.
Reports will be generated regarding Diabetes with
updated patients’ spatial distribution; which allows
to efficiently reallocating medical resources. Most
importantly, spontaneous response in emergency
case can be delivered to citizens; through medical
sensing and GPS signal sent to emergency units.
The proposed framework is an efficient solution
to the above mentioned problems. Having mobile
computing as an interface with citizens permits an
interactive communication with them; storing their
personal health records updated. The proposed
framework promises a significant enhancement in
the health resources allocation and utilization
problem, as well as healthcare service provision.
This section is organized as follows. Sub-section
3.1 lists the intended beneficiaries and contributors.
Sub-section 3.2 illustrates the architecture and sub-
section 3.3 details the modules of the framework.
4.1 Beneficiaries
The beneficiaries and contributors in the proposed
framework are intended to be:
Diabetic Patients (or rather the citizens enrolled
in the governmental medical insurance): Patients
store their own health-related data through the
mobile application; the Personal Health Record
(PHR). Patient can enter all info related to his
disease management; such as daily blood glucose
level, regular check-ups.
Medical care institutions: Medical care
institutions maintain and manage PHRs entered
by patients; updating them with medical
information to be stored as Electronic Health
Records (EHR) in the cloud and shared with
professionals specialized in Diabetes. Diabetes
professionals can import data in EHRs; such as
x-ray photos or laboratory tests.
Emergency units: Emergency units receives
location signal from the citizen mobile GPS, in
case of emergency. Emergency units can then
handle the case giving spontaneous response.
Governments: Analyzing data with high refresh
rates gathered from the country population,
generated reports about diabetes distribution help
in efficient resources reallocation.
NGOs: Acquiring any kind of reports, it could be
generated out of the large and well-analyzed data
collected. This can help such organizations to
effectively direct their resources for better
serving the community.
4.2 Architecture
The proposed framework is based on Mobile Cloud
Computing. The framework architecture is
illustrated in Figure 1.
Cloud Computing has made a revolution on the
computing world by the non-traditional mechanisms
in computing. Governments, getting use of the
widely know advantages of cloud infrastructure; can
efficiently provide their healthcare services at lower
cost computing infrastructures.
On the other side, smart phones and tablets are
considered as the representative for the various
mobile devices as they have been connected to the
Internet with the rapidly growing of wireless
network technology. Ubiquity and mobility are two
major features in the next generation network which
provides a range of personalized network services.
Figure 1: Framework architecture.
4.3 Modules
The framework is composed of two modules sets;
one on the mobile side for citizens’ functionalities
and the other on the cloud side for governments and
healthcare sector functionalities. Interaction of
different modules is depicted in Figure 2.
The description of modules is as follows:
Mobile Application: is the citizen’s interface for
PHR data. The application is deeply analyzed to
include the major diseases with their types,
different degrees of severity and medications. It
is meant to be the data input channel of citizens.
Such mobile application is an effective
companion for the patient; providing updated
ANovelMobile-CloudbasedHealthcareFrameworkforDiabetes
265
guiding about his health case. In case of
emergency, a signal is sent to the emergency
units, allocating his position by GPS, for his life
saving. Medical sensors for glucose level are
integrated with the smart phone.
Privacy Control module: Health information
requires extra protection; as its disclosure can
have serious repercussions in the content owner’s
private and professional life. Husky eHealth 2.0
(Levy, Sargent and Bai, 2011) is a prototype of
healthcare social network system with a trust-
aware tag-based privacy control scheme is
embedded in the framework for privacy control.
The scheme protects private information from
unauthorized access using both tagging and trust
ratings information.
Cloud Data Repository: a repository of data
stored on the cloud representing a real updated
PHR for diabetics to be used for effective data
mining analysis.
EHRs module: a system of EHRs updated by
medical and health professionals and will be used
for effective data mining analysis.
Medical Image Archives Management: Server-
side part of the service realizes medical image
archives management, pre-processing and
rendering of these images in the cloud. Main
functions of client-side are interaction with the
end-user and visualization of rendered
information (Vazhenin, 2012).
Emergency Case Indicator: Emergency units
receives location signal from the citizen mobile
GPS, in case of emergency. Emergency units can
then handle the case; giving spontaneous
response.
Data mining Module: Health records and
feedbacks from citizens are processed through
smart data mining techniques and knowledge
discovery algorithms. Generated reports about
Figure 2: Framework modules.
diabetes distribution are the input in resources
management and reallocation. Another set of reports
help NGOs to effectively direct their resources for
better serving the communities.
Medical Resources allocation module: Resources
are easily reallocated based on the reports for
better service provision and cost cutting.
5 IMPLEMENATATION
To illustrate the benefit of the proposed framework,
an end-to-end scenario is discussed passing through
the entire components of the proposed framework.
First, we show how data is acquired from the
patients through the developed mobile application
and then stored on the Cloud. Second, we describe
the knowledge and conclusions that we can extract
from the acquired data through the data mining
techniques adopted on the Cloud. This scenario has
been simulated using Junosphere, a cloud simulator.
This simulation will clearly demonstrate the benefits
provided by the proposed framework to both
diabetic patient and government decision makers.
Results are then demonstrated and the applicability
of the framework in different countries is discussed.
5.1 Mobile-based Application
for Diabetes Management
The mobile-based application (Salama and Shawish,
2013) was developed to provide a comprehensive set
of functionalities that help the diabetic to easily and
smartly supervise his disease on daily basis. These
functionalities cover a careful follow-up of the
patient’s glucose level, nutrition, and medications. It
also keeps him always connected with his physician,
up-to-date with new diabetes treatments, in addition
to many other services that make the patient able to
fully manage his diabetes.
Using this application, the patient can add his
log; including result of daily glucose level,
medications information, and display all the past
results. The results are analyzed and shown as
visualized graphs. The application also embeds an
updated guide for the patient, to plan healthy eating,
and schedule exercises, gathering all new
information about the disease from different
specialized portals. Figure 3 is a sample of a patient
log using the developed mobile application.
The mobile-based Diabetes management
application requires logging of the following data
(Salama and Shawish, 2013):
Patient basic info; name, date of birth, gender,
smoking status, other medical records, other
genetic records, place of residence
Daily records of blood glucose figures, insulin
HEALTHINF2014-InternationalConferenceonHealthInformatics
266
dosages, records of medications.
Food and exercises logs
BMI (Body Mass Index), BMR (Basal Metabolic
Rate), and Calories Burned Calculator
Records of health downs
Current location, through GPS
Figure 3: Interface of Patient log.
Having the data items logged on regular basis in an
easy manner, the patient will be able to perform his
medical follow-up correctly and his medical history
will also be kept. On the other hand, such data will
be securely saved on the Cloud side to be further
processed through knowledge extraction techniques;
as clarified in the below section.
5.2 Cloud Simulation
The cloud architecture has been simulated using
Junosphere, a cloud simulator (Junosphere, 2013);
leveraging the features of building complete
topologies, connecting physical devices and passing
real traffic to the simulation. A topology is built
using two VMs, two VJX routers with 16 ports (vjx0
and vjx1) that are connected to each other through
port em1 and a Centos (Linux) server. The two VMs
are acting for generating bulk data entry for
simulation purpose and the Centos server is acting as
Figure 4: Cloud simulation.
data repository and reports generator. Mobile
devices are also connected to the built topology by
Junosphere connector. Figure 4 is a screen shot of
the simulated topology.
5.3 Results
Data with high refresh rate, gathered through the
mobile application from the citizens, are processed
through knowledge extraction techniques. The
below information can be straight forward deduced:
Spatial distribution of the patients, medication
needs, and the disease severity.
Daily follow-up of the patients’ status.
While the following knowledge can be easily
concluded:
Real needs, usage and excess of medical
resources.
Efficiency of the healthcare centers locations.
Coverage of the needed physicians
Efficiency of medical campaigns in rural areas
Real needs of diabetes medications and their
availability
Factors causing diabetes and high records based
on location
Taking the diabetes in UK as an example, the
distribution of the diseases in UK is shown in Figure
5 (Diabetes amputation rates show huge regional
variation , 2012), while the distribution of people
with diabetes who received all basic care processes
recommended in the Department of Health’s
standards of care is illustrated in Figure 6 (QiC
Quality in Care, 2012). In fact, the National Audit
Office (NAO) report published in 2012 stated that
only half of the 2.34m people diagnosed with
diabetes received the nine annual basic care
processes. Meanwhile, both maps show how much
efforts and resources should be redirected to other
regions where diabetes is present with high
percentages, instead of being wasted in unneeded
areas. Having accurate information from diabetics
and actual spatial distributions would help
overcoming resources allocation problems.
The extracted information will help the decision
makers and researchers in the following:
Efficient distribution of medical resources, based
on real needs
Smart planning for building healthcare units
Planning for healthcare and medical mobile
campaigns, based on new disease spread
Efficient planning for medications industry, based
on real demand with no excess in production
Smart prediction of similar future patterns for
diabetes
Directing NGOs to put their efforts in the right
places, where the disease needs real effort
Putting research efforts in the right track; in terms
of factors causing the diseases and effective
treatment
ANovelMobile-CloudbasedHealthcareFrameworkforDiabetes
267
Figure 5: Diabetes Distribution UK.
Figure 6: Diabetics who received all basic care processes.
Governments with current strong healthcare systems
are meant to enhance their services by effectively
reallocating resources. The literacy rate, people
awareness, financial conditions, healthcare
investments, population covered under healthcare
insurance policies are much better in these countries.
Meanwhile, this framework represents a cost cutting
in the health sector by implementing efficient
resources allocation and lowering the IT
infrastructure cost. With the advances in their health
sector, better service could be provided; by
implementing enhanced secured EHRs and
emergency indicator.
On the other side, healthcare scenario in
developing countries doesn’t report a good status. A
large percentage of people are residing in rural areas.
The population density is high near big cities. Rural
areas, low density population areas or hilly areas
have shortage of medical facilities with the existence
of primary care centers only, where people rush to
the urban areas for expert medical advice or in case
of emergency. The literacy rate in the developing
countries is very low and the illiteracy rate is very
high in the rural areas. Those people aren’t much
aware of their healthcare needs; they do not follow
the regular checkups and emergent. Hence, being
connected with those people through mobile
application will make their life easier; while the
government will be updated of approach the doctor
in case the situation become their medical condition
and could easily allocate more resources in case of
needs.
5.4 Expected Benefits
The proposed framework presents mutual benefits
for both diabetics and governments. The mobile
application helps the citizens on personal basis
managing their disease case and on-going life easily.
Storing the patient records regularly helps the
physician in correctly prescribing the case. Since the
physician can update the record with more detailed
medical information, the complete medical history
for the patient is, then, stored and kept. The cloud
storage allows easier retrieval from any spot of the
healthcare system along the country.
From governments’ perspective, the use of cloud
is a cutting cost in the IT infrastructures for e-health
services. In addition, the extracted knowledge from
the acquired up-to-date data promises a significant
enhancement in the healthcare related decision
making process. Adding the emergency case
indicator connected to GPS satellites helps in
providing better service and life saving.
5.5 Comparative Evaluation
Comparing the proposed framework with the current
mobile ad cloud-based solutions, we can easily
notice that the proposed one enable the data saved
by the diabetes patient to be share with his physician
opening the chance for more interaction especially in
critical cases. This is in addition to the combined set
of functionalities that are scattered among a group of
separate applications. The proposed framework also
enabled the anonymously collected medical data to
be further analysed on the Cloud, opening a perfect
chance for knowledge discovery based on a high rate
of fresh data and hence smarter decision making.
On the other hand, the available mobile-based
solutions are limited to personal use only without
any external interaction, while Cloud-based
solutions focus on saving healthcare records
especially big sized data like medical images from
hospitals and medical centres without a true effort
the extract knowledge out of this data.
HEALTHINF2014-InternationalConferenceonHealthInformatics
268
6 CONCLUSIONS AND FUTURE
WORK
The proposed framework smartly combined the
broadly available technologies; smart phones and
Cloud computing; in addition to intelligent data
mining techniques to efficiently provide a smart
healthcare framework for Diabetes.
With the fast and wide availability of smart
phones, the mobile application represent a low cost,
fast, and vigorously tool that help the government to
acquire knowledge from the citizens in a parallel
manner, while saving valuable time and resources. It
has also been recognized as a fast communication
channel to deliver guiding instructions and
spontaneously manage emergencies.
With broad availability, scalability, and huge
storage capabilities, the Cloud, on the other hand,
showed a perfect ability to accommodate the
healthcare system holding tones data of millions of
concurrent users. The data mining techniques and
knowledge discovery algorithms have smartly
benefit from the huge amount of fresh data.
Important and useful relations between data have
been deducted allowing an efficient utilization and
reallocation of medical resources, in addition to
predict disease patterns and hence efficiently cope
with them.
Our future work focuses on the development of
interactive mobile applications covering more
endemic diseases with the appropriate connection
with the cloud-bases healthcare system.
REFERENCES
Retrieved 2012, from glucose buddy:
http://www.glucosebuddy.com/
Retrieved 2013, from QiC Quality in Care:
http://www.qualityincareblog.org.
Retrieved 2012, from IBG Star: http://www.ibgstar.us.
Retrieved April 2013, from Organization of economics co-
operation and development: http://www.oecd.org/
Retrieved 2013, from Diabetes.co.uk: http://
www.diabetes.co.uk/
Retrieved 2013, from BBC - Health: Diabetes:
http://www.bbc.co.uk/health/physical_health/condition
s/in_depth/diabetes/index.shtml.
Retrieved 2013, from Diabetes matters: https://
www.diabetesmatters.co.uk/
bglumon. (2012). Retrieved 2012, from Apple Store:
http://itunes.apple.com/us/app/bglumon-blood-
glucose-monitor/id353118307?mt=8.
Botts, N., Thoms, B., Noamani, A., and Horan, T. (2010).
Cloud Computing Architectures for the Underserved:
Public Health Cyberinfrastructures through a Network
of HealthATMs. 43rd Hawaii Int. Conf. on System
Sciences (HICSS). Hawaii, USA.
Diabetes amputation rates show huge regional variation .
(2012). Retrieved 2013, from BBC News:
http://www.bbc.co.uk/news/health-17270379.
Doukas, C., Pliakas, T., and Maglogiannis, I. (2010).
Mobile healthcare information management utilizing
Cloud Computing and Android OS. Annual Int. Conf.
IEEE Engineering in Medicine and Biology Society
(EMBC). Buenos Aires, Argentina.
Hamdan, S. (2011). Rapid Increase of Diabetes Strains
Middle East’s Health Agencies. New York Times, The
International Herald Tribune Section.
Junosphere. (2013). Retrieved November 2013, from
Juniper Networks: http://www.juniper.net.
Koufi, V., Malamateniou, F., and Vassilacopoulos, G.
(2010). Ubiquitous access to cloud emergency medical
services. 10th IEEE Int. Conf. on Information Tech.
and App. in Biomedicine (ITAB). Corfu, Greece.
Levy, K., Sargent, B., and Bai, Y. (2011). A trust-aware
tag-based privacy control for eHealth 2.0. Information
technology education. West Point, NY, USA.
Löhr, H., Sadeghi, A., and Winandy, M. (2010). Securing
the e-health cloud. 1st ACM Int. Health Informatics
Symposium (IHI’10). Arlington, VA, USA.
Mohammed, D., and Fiaidhi, J. (2010). HCX: A
Distributed OSGi Based Web Interaction System for
Sharing Health Records in the Cloud. IEEE/WIC/ACM
Int. Conf. on Web Intelligence and Intelligent Agent
Technology (WI-IAT). Toronto, Canada.
On Track. (2012). Retrieved 2012, from Android Market:
https://play.google.com/store/apps/details
?id=com.gexperts.ontrackandhl=en.
Salama, M., and Shawish, A. (2013). A Comprehensive
Mobile-Based Companion for Diabetes Management.
7th WSEAS European Computing Conf. Dubrovnik,
Croatia: WSEAS.
Vazhenin, D. (2012). Cloud-based Web-service for Health
2.0. Int. Conf. on Human-Centered Computer
Environments. Aizu-Wakamatsu, Japan.
Vree. (2012). Retrieved 2012, from Apple Store:
http://itunes.apple.com/us/app/vree-for-
diabetes/id355923059?mt=8.
Wooten, R., Klink, R., Sinek, F., Bai, Y., and Shar, M.
(2012). Design and Implementation of a Secure
Healthcare Social Cloud System. 12th IEEE/ACM Int.
Symp. Cluster, Cloud and Grid Computing (CCGRID
2012). Ottawa, Canada.
ANovelMobile-CloudbasedHealthcareFrameworkforDiabetes
269
