Content uploaded by Chung-Horng Lung
Author content
All content in this area was uploaded by Chung-Horng Lung on Nov 01, 2017
Content may be subject to copyright.
Internet of Things: Remote Patient Monitoring
Using Web Services and Cloud Computing
Junaid Mohammed
Dept. of Systems and
Computer Engineering
Carleton University
Ottawa, Canada
junaidmohammed@cmail.carleton.ca
Abhinav Thakral
Dept. of Systems and
Computer Engineering
Carleton University
Ottawa, Canada
sunnythakral@cmail.carleton.ca
Adrian Filip Ocneanu
Dept. of Systems and
Computer Engineering
Carleton University
Ottawa, Canada
adrianocneanu@cmail.carleton.ca
Colin Jones
Dept. of Systems and
Computer Engineering
Carleton University
Ottawa, Canada
colinjones@cmail.carleton.ca
Chung-Horng Lung
Dept. of Systems and
Computer Engineering
Carleton University
Ottawa, Canada
chung.lung@sce.carleton.ca
Andy Adler
Dept. of Systems and
Computer Engineering
Carleton University
Ottawa, Canada
adler@sce.carleton.ca
Abstract -- The focus on this paper is to build an Android
platform based mobile application for the healthcare
domain, which uses the idea of Internet of Things (IoT) and
cloud computing. We have built an application called ‘ECG
Android App’ which provides the end user with visualization
of their Electro Cardiogram (ECG) waves and data logging
functionality in the background. The logged data can be
uploaded to the user’s private centralized cloud or a specific
medical cloud, which keeps a record of all the monitored
data and can be retrieved for analysis by the medical
personnel. Though the idea of building a medical application
using IoT and cloud techniques is not totally new, there is a
lack of empirical studies in building such a system. This
paper reviews the fundamental concepts of IoT. Further, the
paper presents an infrastructure for the healthcare domain,
which consists of various technologies: IOIO
microcontroller, signal processing, communication protocols,
secure and efficient mechanisms for large file transfer, data
base management system, and the centralized cloud. The
paper emphasizes on the system and software architecture
and design which is essential to overall IoT and cloud based
medical applications. The infrastructure presented in the
paper can also be applied to other healthcare domains. It
concludes with recommendations and extensibilities found
for the solution in the healthcare domain.
Keywords: Internet of Things, cloud computing,
healthcare applications, system and software infrastructure
I. INTRODUCTION
The Internet of Things (IoT) is the next paradigm shift,
where sensors are connected to the Internet, which collect
data for analysis to make our planet more instrumented,
interconnected and intelligent [1]. A typical person carries
on average one or two mobile devices nowadays. Hence,
by leveraging the increasing presence of mobile devices
the cost of equipment can be reduced significantly in
many industries.
New services can emerge to address society challenges
such as remote health monitoring for elderly patients.
A. Motivation
The ECG Android application presented in this paper
focuses on the health care domain of IoT. With the
advancements in embedded information and
communication technologies, we can provide intensified
healthcare support of senior citizens at homes and
retirement homes. This type of technology would be
helpful to be providing ECG monitoring facility to senior
citizens, athletes and common people. By providing the
facility to use these technologies in the home, citizens
would be able to live independently for a longer period of
time, helping to reduce costs of medical equipment and
the need for additional caregiver resources in the process.
With an increase in the speed and volume of health
related data, we need new technologies that would help in
analyzing the data and will enable in predicting the illness
of a patient and will further help the caregiver to make
better decisions.
Healthcare is currently facing the challenge of large
amount of data that is unstructured, diverse and growing
at an exponential rate. Data is constantly streamed
through sensors, monitors and instruments in real time
that is faster than the medical personnel can keep up with.
For instance, there are about 16,000 hospitals worldwide,
which collect patient data and 80% of this data is
unstructured [1]. The advanced techniques and high
capacities of cloud computing, processing of a large
amount of data can be performed more efficiently support
big data analytics.
B. Problem Statement
In the healthcare domain of IoT, patients will not have
to make as many trips to the doctor anymore, since they
2014 IEEE International Conference on Internet of Things (iThings 2014), Green Computing and Communications (GreenCom
2014), and Cyber-Physical-Social Computing (CPSCom 2014)
978-1-4799-5967-9/14 $31.00 © 2014 IEEE
DOI 10.1109/iThings.2014.45
257
2014 IEEE International Conference on Internet of Things (iThings 2014), Green Computing and Communications (GreenCom
2014), and Cyber-Physical-Social Computing (CPSCom 2014)
978-1-4799-5967-9/14 $31.00 © 2014 IEEE
DOI 10.1109/iThings.2014.45
256
2014 IEEE International Conference on Internet of Things (iThings 2014), Green Computing and Communications (GreenCom
2014), and Cyber-Physical-Social Computing (CPSCom 2014)
978-1-4799-5967-9/14 $31.00 © 2014 IEEE
DOI 10.1109/iThings.2014.45
256
can upload the collected data from the sensors to the
cloud from the comfort of their home for a doctor or
trained specialist to review. This can be achieved for an
ECG monitoring application on the mobile device, which
will collect the bio-signal data using a micro-controller
and then upload to the cloud for keeping a record of the
unstructured data. This will reduce the waiting time for
the triage at the hospitals and minimize visits, but more
importantly reducing the cost of personnel and
administrative operations. This convenience increases the
quality of life for the patients as they can enjoy other
activities instead of spending time commuting to the
hospital/clinic and waiting in long triage queues. Pattern
recognition and analysis can also be applied in real time
across a large set of data to support things like predicting
heart strokes for cardiovascular patients.
The huge volume of data produced from the sensors is
in an unstructured format, which is very complex to
understand and requires different data storage
mechanisms than the typical database management
system (DBMS). In summary, distributed computing,
cloud computing and faster processors allow the analysis
of this data explosion manageable in order to make
improvements in human life, environment interaction as
well as social connection.
C. Proposed Solution
The idea of building an integrated IoT and cloud based
solution for healthcare applications has been around. For
instance, Lee et al. [2] demonstrated a smart phone based
bio-signal monitoring approach and Hii et al. [3]
presented healthcare solutions using Android devices.
Recently, Authors in [4] also described a systematic
review of various mobile healthcare approaches. Fong and
Chung [5] presented a mobile cloud-based ECG
monitoring service.
However, not many reports in the literature discuss the
related system and software engineering (SSE)
technologies, which is practically important, as there is a
empirical need to build a solid infrastructure by applying
SSE technologies to this area. In addition to medical
knowledge, various SSE technologies are involved in IoT
based healthcare applications, including microcontroller
and sensor technologies, signal processing,
communication protocols, system and software design
(using well documented design patterns), DBMS, web
services, data analysis, and cloud techniques. Such an
infrastructure should not only satisfy the basic functional
requirements, but also address some key non-functional
quality requirements, such as performance,
privacy/security, portability, scalability, flexibility, and
cost.
Using the idea of IoT and cloud techniques, this paper
presents a solution to use an IOIO microcontroller board,
which obtains the bio-signal data from a person using
ECG electrodes and sends it to the mobile device
wirelessly using Bluetooth technology. When monitoring
the ECG of the patient, the monitored data associated with
the ECG waves being displayed on the mobile app is
stored in the form of a binary file on the secure digital
(SD) card of the device and the user has the ability to
upload it to a structured query language (SQL) Server
private database. The Filestream and Filetable
technologies present in Microsoft SQL Server [7] 2012
allow the storage of unstructured data. With the proper
hardware components like the IOIO microcontroller and
the ECG electrodes, the solution can monitor the ECG of
a person in any environment at low costs, without having
to purchase any costly ECG monitoring devices.
D. Accomplishments
We have successfully designed, implemented and
tested an integrated infrastructure for the ECG Android
application. The following is a list of main
accomplishments:
• ECG waves visualization on the Android App.:
The ECG waves were plotted by the data
transmitted from the sensors on the IOIO
microcontroller via Bluetooth technology.
• ECG data logging on the Android App: The large
amount of data received from the IOIO
microcontroller is covered to binary file which
would contain all the data received and this file
would be stored on the SD card of the mobile
phone. This feature improves the performance and
scalability.
• An Upload Service: The service uploads the files
on the SD card of a device to a private centralized
cloud using an FTPES secure server. The transfer
of the file takes place using File Transfer Protocol
over the Internet Protocol. We have developed the
ability to store unstructured data on a File system
without causing any form of latency within the
database using filetable and Filestream technology
in Microsoft SQL Server 2012. Multiple devices
can pass ECG data to the server at the same time as
the cloud has multi-user and multi-device support.
• File Compression and Security: The medical data is
stored in a binary format which would be
encrypted and uploaded to the cloud in a secure
manner using FTPES protocol. The format is
optimized for compact storage and faster byte
parsing on the cloud which is later used for
visualization with MATLAB or any signal
processing software.
II. OVERVIEW OF CONCEPTS
The section provides insights about the IOIO
microcontroller, web service and the cloud structure
associated with this framework.
258257257
A. IOIO - OTG Microcontroller
The IOIO-OTG [6] (pronounced “yo-yo-O-T-G”) is
a development board specially designed to add advanced
hardware I/O capabilities to Android devices. The board
provides robust connectivity to an Android device via a
USB or Bluetooth connection and is fully controllable
from within an Android application using a simple Java
API and no embedded programming or external
programmer will ever be needed. The hardware and
software are open source. IOIO uses Eclipse as the same
Integrated Development Environment for Android
development.
IOIO-OTG leverages the USB On-The-Go
specification to connect as a host or an accessory. There
are several ways to connect the IOIO to a Java
application. For an application running on an Android
device, the IOIO-OTG will act as a USB host and supply
charging current to the device. Applications running on a
Windows, Linux or OSX machines, the IOIO-OTG will
assume device mode and present itself as a virtual serial
port. When in the device mode, the IOIO-OTG can be
powered by the host. By connecting a USB Bluetooth
dongle will cause the IOIO-OTG to show up as a
Bluetooth serial connection and makes it wireless.
The board includes a JST connector for attaching a Li-
Po battery and there are several pin headers broken out for
voltage and ground access. A trim pot on the board allows
adjusting the charge current used when the IOIO-OTG is
acting as a host. A USB-A to micro-A OTG cable is
included which connects to an android device to the
micro-USB port on the board using the cable that comes
with the Android device.
The process of writing to an SD card consumes battery
power of the mobile device, which must be controlled if
writing will be a frequent task on the mobile application.
Writing algorithms will need to be developed by taking
into account the amount of data being collected and at
what frequency, to optimize the writing frequency to the
SD card and ultimately manage the battery consumption
of the mobile device. The current design focuses on the
functional aspects. Further improvement can be conducted
for this area.
The client server architecture is used for storing the
data produced by the mobile application (client) onto the
server. By having a separate server for storing the data on
the SD card of the device, the design has enabled a
controlled environment in which we can address privacy
concerns and introduce encryption for data security.
The processing of converting the binary output file to a
meaningful object file is done on the server side. The
main purpose of doing this conversion is to enable the
analyst to perform analysis on the data acquired from the
mobile application.
The IOIO microcontroller performs the analog to
digital conversion of the ECG bio signal. The bio-signal
data is initially being measured and streamed on the IOIO
microcontroller board's internal buffers and is transmitted
to the mobile device using a Bluetooth channel. The data
received on the device is stored temporarily on the SD
card of the mobile device. When the data recording is
complete, the bio signal data can be uploaded to the
centralized cloud server by the end user. The data analyst
will have the ability to download the data from the
centralized cloud for analysis. The application will
serialize the data onto the SD card in a binary format, so
that the user can only read and make sense of the data if
they have access to the de-serialization algorithm. On the
cloud server, user authentication and authorization
provides a primary level of security and protects the data.
Only the doctors, physicians, sport coaches and any other
privileged viewers will have access to the patient’s or
sport athlete’s bio signal data.
B. Web Services
In IoT, the 'things' have memory shortage, processing
power constraints and also battery consumption issues,
which are all critical concerns for mobile devices.
Therefore, most of the processing is pushed to the server
side, which is following a centralized design. In the three-
tier architecture, which is composed of an application, a
service layer and a database layer, separation of concerns
can be addressed with ease. By allowing an intermediary
web service which abstracts away the authorization,
business logic and backend database implementation, it
makes the application more lightweight and limits the
access to the database.
The web service design is different from a typical
Representational State Transfer (REST) or Simple Object
Access Protocol (SOAP) web service APIs due to the
nature of the unstructured data and the filetable
technology
Since a private cloud server is being used for sensitive
medical information, the data is encapsulated in binary
files and requires minimum data processing. The patient
on the mobile device interfaces with the web service
composed of a File Transfer Protocol (FTP) server, which
is implemented using FileZilla to upload their biomedical
data. On the other hand, the data analyst who needs to
analyze the patient's ECG waves can use the same web
service FTP server to download the required files for
pattern analysis. Instead of the peer to peer (P2P)
communications, there exists a higher level of
communication, method which is machine-to-machine
communication (M2M) in IoT.
FileZilla server was chosen because it is open source
software which is free and it can be installed on Windows
and Linux. The design decision introduces deployment
flexibility as this web service can sit on the cloud, which
is composed of heterogeneous architecture hardware
machines. The FTP server will also be responsible for
interacting with the database server to store and organize
the biomedical output binary files. The interface between
259258258
the FTP server and the database is a synthesized windows
share, which represents the Filetable technology of the
database.
In IoT, the veracity of the data is an important factor
as it raises concerns whether one can trust the data
gathered by sensors or from data entry. The mobile user
can produce 'white lies' about the ECG data being
recorded on the mobile application. Picture the following
scenario, a child - the mobile user, misuses the mobile
application and records their own ECG signals and
uploading it under their grandfather's identity. When the
doctor analyzes the grandfather's ECG signals, they will
be stumped about the results, as the data is the child's
ECG waves.
Therefore authentication and authorization for the
application is crucial for the validity of the data.
Authentication is the process of identifying a user through
a username and password, whereas authorization is
verifying if that user has permissions or access to perform
the specified operation.
FileZilla server serves the purpose of processing the
multiple authentication and authorization requests from
the mobile devices through the FTP and binary file
storage requests from the mobile clients. Once
authentication is successful, the binary files will be
uploaded using the FTP to the home directory of which
the FTP server is pointing to. The mobile application
contains the username and password, which would be
encrypted to prevent unauthorized access to the server and
better control of the veracity of the biomedical data.
C. Cloud Research and Analysis
The data located in one central location rather than
being distributed apart in different places provides higher
feasibility and data security. Since, it is an ethical
requirement to protect the critical medical data of
individual’ bio signals, hence the centralized architectural
design pattern was chosen for the ECG Android app. In
our architectural design, the data monitored for all the
patients will be stored in one centralized location, which
will be separated through a unique identifier to identify
the data for different individuals Since all the data are
stored in one place, it will be easy to query the database
and perform data analysis out of the combined data.
The following are some advantages and disadvantages
of centralized architectural design pattern:
Advantages:
• The data are easily placed in the server.
• There is an effective use of space for the storage of
the data within the cloud.
• All the related data are kept together.
• Data redundancy are avoided.
• It is a uniform service provided to all users.
• The data security is improved in comparison to
decentralized system.
Disadvantages:
• The data may be too distant from the user for
adequate service.
• There is an overhead of maintaining user
authentication for the different users
The unstructured binary files uploaded by mobile users
via a FTP server are stored in a filetable. A filetable is a
database table, which contains file metadata columns and
stores the files to a new technology file system (NTFS)
system outside of the database, which reduces overhead
and improves database performance. The filetable
technology was released by Microsoft in SQL Server
2012, which is actually built upon the Windows
FileStream feature exposed in SQL Server 2008. The
FileStream feature was introduced for binary large objects
(BLOBs) in order to perform full text searches and byte
streams on the BLOBs instead of complex SQL
statements since the contents of the data is not structured
in a row and column format.
The filetable exposes the FileStream folder hierarchy
through a synthesized windows share. The filestream,
which contains all the files contents, has a reference in the
filetable in the form of the varbinary column data type.
The FTP server copies the uploaded binary files to the
windows. The filetable maintains synchronization of the
files and its file metadata. The filetable can be accessed
via transactional access with SQL or non-transactional
access via common run time (CLR) scripts and windows
APIs.
The centralized storage solution can be scalable as a
coordinate node can sit as the interface between the
mobile devices and the storage mechanism and perform
load balancing on multiple back-end database servers.
Therefore, this solution can be expanded to cloud farms
with backup support and high available disaster recovery
strategies.
The biomedical data that is being dealt can be
electrocardiogram, pulse, temperature, blood pressure and
respiratory rate which reveal and health conditions of the
patient. Therefore, the data that are on the application is
highly confidential and must be stored accordingly so that
only the authorized viewers have access to the data. Data
encryption, authentication and authorization practices are
deployed in the biomedical implementation of the IoT
application.
III. DESIGN AND IMPLEMENTATION
The Architecture of the entire end-to-end solution of
the system is based on the layered architectural design
pattern. This pattern divides the system into different units
called layers. These layers group modules, which provide
a cohesive set of services. The main idea behind this
pattern is to have separation of concerns between the
portions of the system, which enables developing and
260259259
maintaining them separately, therefore ma
k
independent of each other.
The end-to-end system architecture for
p
roject involves the hardware, the mobile
a
the cloud. These act as the three major la
y
Patient Monitoring System. Hence, the sy
s
major layers, the hardware layer, applicatio
n
cloud layer. The application has three sub l
a
follows: Service layer, Platform Applica
t
The File Transfer and Writing layer.
Figure 1 shows how the multiple layer
s
architecture interact with one another. The
h
contains the IOIO microcontroller and
s
collects the bio signal data and this data is
the Bluetooth channel on the microco
n
Application layer on an Android device.
The Application layer contains three su
b
the layer itself. The Service Layer is the t
o
application layer, which interacts with the
h
The ECGService is present within the
which is responsible for retrieval of the
b
from the hardware layer and storing the da
t
within ECGModel, which performs the
w
data.
Figure 1: Layered Architecture of Patient Monit
o
The ECGService is a service ru
n
b
ackground of the application, the Main
p
resent within the Platform Application L
a
service. The Main Activity works in the sa
m
an Android App as a main program wor
k
p
rogram. The Platform Application Laye
r
k
ing the portion
this IoT based
a
pplication and
y
ers within this
stem has three
n layer and the
a
yers named as
t
ion layer and
s
in the system
h
ardware layer
s
ensors, which
transmitted by
n
troller to the
b
layers within
o
p layer in the
h
ardware layer.
Service layer,
b
io signal data
t
a in the buffer
w
riting of the
o
ring System
n
ning in the
Activity class
a
yer starts this
m
e manner for
k
s in a regular
r
also contains
the Model, View and the Control
mobile application considers
p
rolonging the lifetime of the se
n
by performing ECG sampling at
every hour or every 10 minut
e
recording.
The Main Activity creates
contains all the background logi
c
View & Controller are observer
s
design pattern, which observe
a
ECGmodel. The View is basica
l
the a
p
p, which the end user inte
r
user on the interface are detecte
d
p
erforms action on the unde
r
ECGModel. Likewise, the contr
o
view according to the data chang
e
The ECGModel contains the
b
y in a timely manner by the E
C
present in the File Transfer and
data from the ECGModel and w
r
file. The ECGModel using the Fi
l
the File Transfer and Writing
identification of the file speci
Upload Manager in this layer
transferring the file from the an
server on the cloud side.
The FTP Server present in th
e
file from the FTP client and th
e
store this file into the File Tabl
e
b
y the database of the cloud laye
r
Figure 2: ECG Waveform Visualizat
i
Illustratio
n
of the Android App. The
energy efficiency and
n
sors and mobile devices
regular intervals such as
e
s instead of continuous
the ECGModel, which
c
of the application. The
s
based on the Observer
a
nd changes made to the
l
ly the User Interface of
r
acts with. Actions of the
d
by the controller, which
r
lying data within the
o
ller will also update the
e
s within the ECGModel.
data, which is provided
C
GService. A File Writer
Writing Layer reads the
r
ites this data to a binary
l
e Tagging mechanism in
Layer does the unique
fic to the device. The
is the FTP client for
droid device to the FTP
e
Cloud layer receives the
e
server is responsible to
e
technology maintained
r
.
i
on on a Mobile Device: an
n
261260260
Since we are working with a classic producer and
consumer problem, where the micro-controller is
producing voltage values that we need to display on the
mobile device and stores it on the SD card on the mobile
device. The Pool Allocation real time design pattern [8]
was used to ensure that none of the voltage values coming
from the producer are dropped, as the physician or doctor
who will be analyzing the data should not be missing
critical data that could lead to saving lives.
The buffer recycling between the availableList and
writeList can be seen below:
Figure 3: Pool allocation pattern
As can be seen in Figure 3, the whole system contains
four buffers at any moment of time, but more importantly,
if one was to observe the Available Linked List, one
would observe on average 1 or 2 buffers as the producer
fills up buffers with data for the writer consumer to
consumer. The pool allocation manager will ensure that
there will be empty buffers to fill for the producer so that
no data is lost since the producer does not have an infinite
queue for the data to queue up. In essence, this design
pattern allows the management between performance and
memory management. The pool allocation manager can
also remove buffers if it notices that some buffers are not
being utilized.
The real cloud aspect is present when we introduce a
controller which will route the mobile device requests
from the different users to different server side nodes
based on the availability and demand. These nodes offer
their resources to process the request and can be increased
or decreased in quantity depending on the demand.
Another challenge with medical data is the confidentiality
of the data. There are three main types of cloud services
which are public, private and hybrid types.
Public cloud service is offered by Google, Amazon
and other main providers; however, hospitals and medical
clinics cannot afford to store their data on these clouds as
they are not Health Insurance Portability and
Accountability Act (HIPAA) compliant, which means that
their implementations do not comply to the rules and
regulations listed under the HIPAA. However, Microsoft's
cloud service, Azure, is HIPAA compliant. Therefore, the
current Microsoft SQL Server implementation can be
deployed to an Azure platform at a private datacenter to
be fully functional.
On the other hand, a private cloud service offers the
same functionality, except the owner of the cloud must
own and maintain its own infrastructure at a highly secure
location. Therefore, Google, Amazon and other main
public cloud providers will not legally own the data and
cannot view the data. Maintaining an infrastructure
increases the cost of owning the private cloud for the
hospital and medical clinics, but more importantly the
infrastructure is not elastic. Elasticity is ability for a
system to adapt to changes in demand and resources. For
example, if the hospital requires more processing power
within a short notice, it will be very difficult for them to
achieve this.
It was found that the best choice for remote patient
monitoring currently is to use a hybrid approach where
hospitals will still heavily rely on their own private cloud
service, but they will use a public cloud offering
occasionally. The public cloud offering will be HIPAA
compliant and will be considered when the incoming
workload on the privately maintained cloud is too heavy
to maintain. Therefore the hospitals can then turn to the
big providers for assistance, when they see a spike in
remote patient monitoring and release the resources back
to the providers when the demand decreases.
The block diagram in Figure 4 is the overall
architectural design for the cloud layer. The figure
illustrates the flow and follows the Pipes and Filters
design pattern [9]. The figure explains the inputs and
outputs of each processing element and how the buffering
is done in the Service Broker message queue [7], which is
a key aspect of the cloud layer.
262261261
Figure 4: Cloud architecture design
The Service Broker supports and processes multiple
mobile upload requests at the same time using
asynchronous message queues. The filetable technology
allows the storage of huge amounts of incoming
encrypted binary data files in the NTFS windows share.
The NTFS storage mechanism can be expanded to the
desired amount of storage capacity as there is no limit to
how much data can be stored as long as there is hardware
support. The data is being stored outside of the database,
therefore the buffer pools of the database are not being
used up, hence improving performance. The CLR scripts
parse the binary files for user and file information and
store multiple mobile user data and file metadata into
tables. The tables are optimized for faster SQL queries for
the users accessing the data such as physicians or medical
technicians.
IV. CONCLUSION AND RECOMMENDATIONS
The main objective of this paper was to build an
Android Application in the healthcare domain using the
idea of IoT and cloud computing. IoT is a technology that
is having major impacts in many different domains.
Specifically health care will be greatly benefited from this
technology in the future. The paper focused on monitoring
ECG waves using the Android platform.
The concept of IoT and cloud for medical applications
has been proposed in the field. However, there is a lack of
empirical studies of end-to-end solutions The proposed
solution is this paper involve various technologies,
including microcontroller, signal processing for ECG
waves, communication protocols and system design to
support private and secure data transfer, system and
software technologies for large amount of data storage
and reliable transfer using proven software design
patterns, DBMS, web services, and cloud techniques.
An infrastructure has been built based on those
aforementioned technologies for ECG wave monitoring.
But there is a good potential of extensibility for the
infrastructure, as one could monitor additional vital signs
and perform analytics on the data collected to make
predictions of health conditions of individuals, besides
expanding to other platforms.
The current system design can be further reengineered
for performance improvement. With the help of additional
sensors on the board attached to the IOIO-OTG, more
number of vital signs could be monitored for an
individual. The IOIO-OTG is equipped with 46 I/O pins.
More sensors can be attached to the different I/O pins on
the microcontroller for gathering data from different
sources. The data associated with different vital signs will
be transmitted to the mobile device by the same Bluetooth
channel using the Bluetooth dongle attached to the board.
The data received on the mobile device can be written to
different files. Each file can be given a unique name,
depending on the data source (or the vital sign) that is
written to that particular file. With additional vital signs
being monitored, the User Interface of the App can be
updated accordingly to display additional vital signs data
for the user.
Typically an ECG monitoring machine at hospitals
costs $1,000 to $2,000 which are expensive. They are
proprietary and require a tedious process of retrieving the
data stored on those machines. They also have a weak
user interface with minimal buttons which is not usable by
a typical patient or personnel. With technology being
more accessible, microcontrollers can replace the
proprietary heavy bricks and allow a typical person with a
smart phone to monitor their ECG with an appealing easy-
to-use user interface. They can send their data files to
their physician for analysis with the tap on their screen -
for less than $50 (cost of IOIO microcontroller). This will
also eliminate waiting times in hospitals and will provide
an easy facility to senior citizens to monitor their own
ECG and derived vital signs like respiratory rate; oxygen
levels etc. - as it may be hard for them to travel to
hospitals for diagnostic purposes.
The infrastructure presented in this paper can be
applied to other medical related applications. One
possible example is that athletes can also use this app, as
they can perform their own recording sessions after
intense workouts by gathering their data over a period of
time and sending their medical data to a medical
professional or personal trainer/coach. With all the data
available for the professional they would be able to judge
how the athlete has been performing over a period of time
and how the training program is affecting the athlete’s
vital signs over a period of time, and potentially their
performance.
Physicians or trainers can make use of the data
collected for a long period of time for different
applications for further data analysis. For instance, a
263262262
physician can perform pattern recognition against the
ECG waves using third party software tools to discover
correlations or any causation. The binary file that has
been uploaded to the cloud can be analyzed using
MATLAB, or any software which would take in data files
and support an analysis to predict any dangerous health
conditions that will arise based on the current projection
or the behavior of the recorded bio-signals.
The IOIO microcontroller was simply used to rapidly
develop a proto-type to prove that this system is possible.
The next steps are to use the contact-less sensors along
with the next generation micro-controller which is the size
of your thumbnail.
REFERENCES
[1] Healthcare Data Challenge, available on line: http://www-
03.ibm.com/press/us/en/pressrelease/40788.wss, Last
accessed Jan. 2014.
[2] B..G Lee, W.Y. Chung, “A Smartphone-based Driver Safety
Monitoring System Using Data Fusion”, Sensors, 12, 2012,
pp. 17536–17552.
[3] P. C. Hii,, S.C. Lee, T. H. Kwon, W. Y. Chung, “Smart
Phone Based Patient-Centred Remote Healthcare Monitoring
Application in WSN”, Sens. Lett., 9, 2011, pp. 791–796.
[4] L. C. Jersak, A. C. da Costa, D. A. Callegari, “A Systematic
Review on Mobile Health Care”, Tech. Report 073,
Faculdade de Informática PUCRS – Brazil, May 2013.
[5] E.-M. Fong and W.-Y. Chung, “Mobile Cloud-Computing-
Based Healthcare Service by Noncontact ECG Monitoring”,
Sensors, 13(12), Dec. 2013, pp. 16451-16473.
[6] IOIO, available online: https://www.sparkfun.com/
products/11343, Last accessed Jan. 2014].
[7] SQL Server Service Broker, available online:
http://www.sqlservercentral.com/columnists/sSampath/sb00
4.jpg, Last accessed Jan. 2014.
[8] Pool Allocation Pattern, available online:
http://www.informit.com/articles/article.aspx?p=30309&seq
Num=3, Last accessed Jan. 2014.
[9] Software Architectures, available online:
http://www.csee.wvu.edu/~ammar/CU/swarch/lecture%20sli
des/slides%204%20sw%20arch%20styles/supporting%20sli
des/SWArch-4-PipesandFilter.pdf, Last accessed Jan. 2014.
264263263
