See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/49628402

Mobile Healthcare Information Management Utilizing Cloud Computing and

Android OS

Article in Conference proceedings: ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference ·
August 2010
DOI: 10.1109/IEMBS.2010.5628061 · Source: PubMed

CITATIONS READS
310 13,140

3 authors:

Charalampos Doukas Thomas Pliakas

FBK CREATE-NET CloudStreet Oy
87 PUBLICATIONS 2,729 CITATIONS 24 PUBLICATIONS 520 CITATIONS

SEE PROFILE SEE PROFILE

Ilias Maglogiannis
University of Piraeus
437 PUBLICATIONS 6,398 CITATIONS

SEE PROFILE

All content following this page was uploaded by Ilias Maglogiannis on 20 May 2014.

The user has requested enhancement of the downloaded file.

32nd Annual International Conference of the IEEE EMBS
Buenos Aires, Argentina, August 31 - September 4, 2010

Mobile Healthcare Information Management utilizing Cloud

Computing and Android OS
Charalampos Doukas, student Member IEEE, Thomas Pliakas, student Member IEEE and Ilias
Maglogiannis, Member, IEEE

Abstract—Cloud Computing provides functionality for Computing and Android Operating System (OS) [2].
managing information data in a distributed, ubiquitous and
pervasive manner supporting several platforms, systems and II. MOBILE HEALTHCARE SYSTEMS AND CLOUD
applications. This work presents the implementation of a COMPUTING
mobile system that enables electronic healthcare data storage,
update and retrieval using Cloud Computing. The mobile Several studies have demonstrated that the limited access
application is developed using Google’s Android operating to patient-related information during decision-making and
system and provides management of patient health records and the ineffective communication among patient care team
medical images (supporting DICOM format and JPEG2000 members are proximal causes of medical errors in healthcare
coding). The developed system has been evaluated using the ([8], [9]). Thus, the pervasive and ubiquitous access to
Amazon’s S3 cloud service. This article summarizes the
implementation details and presents initial results of the system healthcare data is considered essential for the proper
in practice. diagnosis and treatment procedure. Cloud Computing is a
model for enabling convenient, on-demand network access
I. INTRODUCTION to a shared group of configurable computing resources (e.g.,
networks, servers, storage, applications, and services) that
M OBILE healthcare systems focus towards achieving two
specific goals: the availability of e-health applications
and medical information anywhere and anytime and the
can be rapidly provisioned and released with minimal
management effort or service provider interaction. This
invisibility of computing [1]. Mobile pervasive healthcare cloud model promotes availability and is composed of five
technologies can support a wide range of applications and essential characteristics, three service models, and four
services including mobile telemedicine, patient monitoring, deployment models. The major characteristics of Cloud
location-based medical services, emergency response and Computing can be summarized into the following [11]: (A)
management, personalized monitoring and pervasive access On-demand self-service. A consumer can unilaterally obtain
to healthcare information, providing great benefits to both access to computing capabilities, such as server computing
patients and medical personnel ([1], [10]). The realization time and/or network storage, as needed automatically
however of health information management through mobile without requiring human interaction with each service’s
devices introduces several challenges, like data storage and provider; (B) Broad network access: Resources are available
management (e.g., physical storage issues, availability and over the network and accessed through standard
maintenance), interoperability and availability of mechanisms that promote use by heterogeneous thin or thick
heterogeneous resources, security and privacy (e.g., client platforms (e.g., smart phones); (C) Resource pooling:
permission control, data anonymity, etc.), unified and The provider’s computing resources are pooled to serve
ubiquitous access. One potential solution for addressing all multiple consumers using a multi-tenant model, with
aforementioned issues is the introduction of Cloud different physical and virtual resources dynamically
Computing concept in electronic healthcare systems. Cloud assigned and reassigned according to consumer demand.
Computing provides the facility to access shared resources Examples of resources include storage, processing, memory,
and common infrastructure in a ubiquitous and pervasive network bandwidth, and virtual machines; (D) Rapid
manner, offering services on demand over the network to elasticity: Resources can be rapidly and elastically
perform operations that meet changing needs in electronic provisioned, in some cases automatically, to quickly scale
healthcare application. In this context @HealthCloud has out and rapidly released to quickly scale.
been developed; a pervasive healthcare information Given the characteristics of Cloud Computing and the
management system for mobile devices utilizing Cloud flexibility of the services that can be developed, a major
benefit is the agility that improves with users being able to
Manuscript received April 23, 2010. rapidly and inexpensively re-provision technological
Charalampos Doukas is with the University of the Aegean, Samos, infrastructure resources. Device and location independence
Greece (e-mail: [email protected]). enable users to access systems using a web browser
Thomas Pliakas is with the Nokia Siemens Networks, COO Converged
Core, Athens, Greece (e-mail: [email protected]).
regardless of their location or what device they are using
Ilias Maglogiannis is with the University of Central Greece, Lamia, (e.g., mobile phones). Multi-tenancy enables sharing of
Greece (e-mail: [email protected]).

978-1-4244-4124-2/10/$25.00 ©2010 IEEE 1037

resources and costs across a large pool of users thus @HealthCloud application and presents implementation
allowing for centralization of infrastructure in locations with details. The prevalent functionality of the application is to
lower costs. Reliability improves through the use of multiple provide medical experts and patients with a mobile user
redundant sites, which makes Cloud Computing suitable for interface for managing healthcare information. The latter
business continuity and disaster recovery. Security typically interprets into storing, querying and retrieving medical
improves due to centralization of data and increased images, patient health records and patient-related medical
security-focused resources. Sustainability comes about data (e.g., biosignals). The data may reside at a distributed
through improved resource utilization, more efficient Cloud Storage facility, initially uploaded/stored by medical
systems. personnel through a Hospital Information System (HIS). In
A number of Cloud Computing platforms are already order to be interoperable with a variety of Cloud Computing
available for pervasive management of user data, either free infrastructures, the communication and data exchange has to
(e.g., iCloud [13] and DropBox [15]) or commercial (e.g., be performed through non-proprietary, open and
GoGrid [12] and Amazon AWS [14]). The majority of them interoperable communication standards. @HealthCloud
however, do not provide to developers, the ability to create utilizing Web Services connectivity and Android OS
their own applications and incorporate Cloud Computing supports the following functionality:
functionality, apart from Amazon AWS. More information Seamless connection to Cloud Computing storage: The main
on the latter issue is provided in Section V, while the application allows users to retrieve, modify and upload
following section discusses related work in the context of medical content (medical images, patient health records and
mobile pervasive healthcare. biosignals) utilizing Web Services and the REST API [17].
The content resides remotely into the distributed storage
III. RELATED WORK elements but access is presented to the user as the resources
are located locally in the device (see Figure 1e).
The application of mobile devices for pervasive Patient Health Record Management: Information regarding
healthcare information management has already been patient’s status, related biosignals and image content can be
acknowledged and well established ([6], [7]). Authors in [3] displayed and managed through the application’s interface
present the benefits of using virtual health records for (see Figure 1a).
mobile care of elderly citizens. The main purpose of the Image viewing support: The DICOM [18] medical image
work is to provide seamless and consistent communication protocol is supported, while the JPEG2000 standard has been
flow between home health care and primary care providers implemented to support lossy and lossless compression,
using devices like PDAs and Tablet PCs. Smart cards and progressing coding and Region of Interest (ROI) coding. The
web interfaces have been used in [4] for storing patient progressive coding allows the user to decode large image
records electronically. The MADIP system [5] is a files at different resolution levels optimizing this way
distributed information platform allowing wide-area health network resources and allowing image acquisition even in
information exchange based on mobile agents. In [10] cases network availability is limited (see examples in Figure
authors present a mobile platform for exchanging medical 1c and Figure 1d). The code for performing wavelet decoding
images and patient records over wireless networks using on mobile devices in [10] has been modified to support the
advanced compression schemes. The majority of the JPEG2000 standard on the Android platform. Image
aforementioned works is based on proprietary architectures annotation is also supported, using the multi-touch functions
and communication schemes and requires the deployment of of the Android OS.
Proper user authentication and data encryption: User is
specific software components. Furthermore, the works focus
authenticated at the Cloud Computing Service with SHA1
mostly on delivering data to healthcare applications and do
[19] hashing for message authentication and SSL [20] for
not address issues of data management and interoperability
encrypted data communication.
issues introduced by the heterogeneous data resources found
in modern healthcare systems. The usage of Cloud V. PROPOSED SYSTEM ARCHITECTURE AND
Computing provides data management and access IMPLEMENTATION DETAILS
functionality overcoming the aforementioned issues as
Figure 2 illustrates the proposed system architecture for
discussed in previous sections. The concept of utilizing
developing and deploying the mobile healthcare applications
Cloud Computing in the context of healthcare information
that utilize Cloud Computing. The main components of a
management is relatively new but is considered to have great
Cloud Computing Service usually are [11] the platform front-
potential [16]. To our best knowledge there is no other work end interface that communicates directly with users and
in the literature utilizing Cloud Computing for providing allows the management of the storage content. The interface
pervasive healthcare information management services on can be a web client or a standalone application. The Cloud
mobile devices. Storage Facilities manages the physical infrastructure (e.g.,
storage elements) and is also responsible for performing
IV. APPLICATION OVERVIEW maintaining operations (e.g., backing up data, etc.)
This section discusses the main features of the
1038
 (a) (b) (c) (d) (e) (f)
Figure 1. Screenshots of the @HealthCloud mobile application: a) Displaying a patient health record, b) illustration of DICOM header extraction, c)
JPEG2000 progressive decoding of a CT scan at frist resolution level (out of five), d) final output of JPEG2000 progressive decoding of a CT scan, e) The
main application interface displaying available files on the Cloud and available operations, f) illustration of the uploading procedure of a file into the Cloud.

The Cloud Platform interface is also connected to the Cloud and heard information data. When JEPG2000 compression
Service module, which handles and queues user requests. is used, the appropriate sub-module decodes the image. The
Finally, the Cloud Infrastructure module manages user communication with the Cloud is performed through an
account, accessibility and billing issues. implementation of Web Services REST API that is
Previous work by authors ([10]) has demonstrated the supported natively by Android. The inherent interoperability
applicability of mobile devices into retrieving medical image that comes with using vendor, platform, and language
data from remote repositories wirelessly utilizing proper independent XML technologies and the ubiquitous HTTP as
content coding (i.e., wavelet compression with region of a transport mean that any application can communicate with
interest support). This work has been now extended to any other application using Web services.
include the functionality of communicating with Cloud Data in Cloud are seamlessly stored and presented to the
Computing platforms and support communication through user as if they reside locally. This means that the Cloud
Web Services. In this context, @HealthCloud has been repository is presented as a virtual folder and does not
developed based on Google’s Android mobile Operating provide the features of a database scheme. In order to
System (OS) [2] using the appropriate software development provide the user with data querying functionality, medical
kit (sdk). Android is a mobile operating system running on records and related data (images and biosignals) are stored
the Linux kernel. Several mobile device vendors already into a SQLite [21] file. SQLite is the database platform
support it. The platform is adaptable to larger and traditional supported by Android. The file resides into a specific
smart phone layouts and supports a variety of connectivity location at the Cloud and is retrieved on the device every
technologies (CDMA, EV-DO, UMTS, Bluetooth, and Wi- time user needs to query data. The query is performed
Fi). It supports a great variety of audio, video and still image locally and the actual location of the data in the cloud is
format, making it suitable for displaying medical content. revealed to the applications. The database file is updated and
Finally, it supports native multi-touch technology, which uploaded into the Cloud every time user modifies data,
allows better manipulation of medical images and generally respectively.
increases the application’s usability.
The Cloud Service client running on Android OS consists I. UTILIZING AMAZON S3 CLOUD SERVICE
of several modules. The Patient Health Record application For the realization of the mobile healthcare information
acquires and displays patient records stored into the cloud. management system the Amazon Simple Storage Service
The Medical Imaging module is responsible for displaying (S3) has been utilized. The main reason for selecting the
medical images on the device. specific Cloud Computing platform is that it is a commercial
service well established and used successfully in several
applications ([22]). It provides users with several
interoperable web interfaces for managing data (SaaS
model) and developers with the ability to create their own
applications for accessing the latter (PaaS model) and is
suitable for managing healthcare information ([23]).

II. INITIAL EVALUATION FROM THE SYSTEM IN PRACTICE

In order to prove the system’s usability, some initial
experiments evaluating the system’s performance have been
conducted. Experiments concern the time needed to transmit
Figure 2. Illustration of the proposed system architecture data to the Amazon S3 Cloud storage service. Due to the
fact that textual data like a patient’s health record or a
It decodes images in DICOM format displaying both image biosignal sequence do not consist of large data files and do

1039
 not require high bandwidth, the presented results involve the healthcare information management system based on Cloud
transmission of medical images. The @HealthCloud Computing and Android OS. The system enables the
application as presented in previous sections has been used management of patient health records and medical images
on a HTC G1 mobile phone running Android OS version (supporting DICOM format and JPEG2000 coding) and
1.6. A number of medical images of different modalities utilizes the Amazon’s S3 Cloud Storage Service.
(MR, CT, PET, OT and Ultrasound) and different file sizes Future work might include improving security by
have been used. The transmission times are displayed in implementing advanced user authentication techniques on
Table I. As indicated, two different wireless network the mobile device (e.g, through voice recognition) and
infrastructure types have been utilized; a WLAN and a deploying the platform in real healthcare environment for
commercial 3G Network. evaluating the system in terms of user acceptability and
TABLE I performance.
TRANSMISSION TIME OF MEDICAL IMAGES USING AMAZON S3 CLOUD
SERVICE AND DIFFERENT NETWORK TYPES
Image Type (encoding) File Size Time (secs) REFERENCES
(MB) [1] Upkar Varshney, “Pervasive Healthcare”, IEEE Computer Magazine
3G Network WLAN vol. 36, no. 12, 2003, pp. 138-140.
Network [2] The Android mobile OS by GoogleTM, http://www.android.com/
OT (24-bit JPEG2000 [3] Sabine Koch, Maria Hägglund, Isabella Scandurra, Dennis Moström,
6.8 42.532 7.894
Lossless Color) “Towards a virtual health record for mobile home care of elderly
CT (Uncompressed) citizens”, presented in MEDINFO 2004, Amsterdam, 2004.
0.528 4.023 2.382
[4] Alvin T.S. Chan, “WWW_smart card: towards a mobile health care
CT (JPEG2000) management system”, International Journal of Medical Informatics
0.102 1.223 0.892
vol. 57, 2000, pp. 127–137.
MR (JPEG Lossless)
0.721 9.738 3.894 [5] Chuan Jun Su, “Mobile multi-agent based, distributed information
platform (MADIP) for wide-area e-health monitoring”, Computers in
PET (JPEG2000 Lossy)
0.037 0.923 0.793 Industry, vol. 59, 2008, pp. 55–68.
Ultrasound (sequence of 10 [6] Khawar Hameed, “The application of mobile computing and
0.487 3.892 3.251 technology to health care services”, Telematics and Informatics, vol.
images, JPEG2000 Lossless)
20, 2003, pp. 99–106.
[7] Eneida A. Mendonça, Elizabeth S. Chen, Peter D. Stetson, Lawrence
The performance of both WLAN and 3G networks can be K. McKnight, Jianbo Lei, James J. Cimino, “Approach to mobile
easily biased by traffic and other network conditions, since information and communication for health care”, International
Journal of Medical Informatics, vol 73, 2004, pp. 631—638.
commercial networks have been utilized in both cases. Also, [8] L.L. Leape, “Error in medicine”, J. Am. Med. Assoc, vol. 272, 1994,
the response time of the Amazon S3 Cloud service can play pp. 1851—1857.
an important role on the total transmission time. However, [9] J.T. Reason, Human Error, Cambridge University Press, Cambridge,
1990.
the acquired results can be considered as indicative since the [10] Maglogiannis I., Doukas C., Kormentzas G., Pliakas T., "Wavelet-
experiments reflect a real case scenario where the specific Based Compression With ROI Coding Support for Mobile Access to
service and commercial wireless networks are utilized in DICOM Images Over Heterogeneous Radio Networks", IEEE
Transactions on Information Technology in Biomedicine, vol.13, no.4,
order to transmit medical data. In addition, the time needed
pp.458-466, July 2009.
to decode and present the specific images used in the [11] George Reese, Cloud Application Architectures: Building
experiments has been measured. For the HTC G1 mobile Applications and Infrastructure in the Cloud, O'Reilly Media,
phone used, the time needed by @HealthCloud to display Paperback (April 17, 2009), ISBN: 0596156367.
[12] GoGrid Storage Services, http://www.gogrid.com
uncompressed CT images at a resolution of 512x512 pixels [13] iCloud, http://www.icloud
was 0.52 sec, compressed CT images with JPEG2000 [14] Amazon Web Services (AWS), http://aws.amazon.com/
coding at a resolution of 512x512 pixels was 4.53 sec. The [15] DropBox, https://www.dropbox.com
[16] Ofer Shimrat, “Cloud Computing and Healthcare”, April 2009 "San
time needed to decode OT images compressed with Diego Physician", pp. 26-29.
JPEG2000 at resolution of 3072x2048 was 21 sec. and 7.5 [17] Leonard Richardson, Sam Ruby, and David Heinemeier Hansson,
sec. for a sequence of 10 ultrasound images of 600x430 Restful Web Services, Paperback, O'Reilly Media, May 2007, ISBN:
0596529260.
pixels. [18] The Digital Imaging and Communications in Medicine (DICOM)
standard, http://medical.nema.org/
III. CONCLUSION [19] US Secure Hash Algorithm 1 (SHA1),
http://www.faqs.org/rfcs/rfc3174.html
The sharing of medical information resources (electronic [20] Eric Rescorla, SSL and TLS: Designing and Building Secure Systems
health data and corresponding processing applications) is a (Paperback), Addison-Wesley Professional (October 27, 2000), ISBN:
key factor playing an important role towards the successful 0201615983.
[21] The SQLite Database Engine, http://www.sqlite.org/
adoption of pervasive or mobile healthcare systems. The [22] Amazon’s AWS Success Case Studies,
concept of Cloud Computing and applications similar to the http://aws.amazon.com/solutions/case-studies/
one presented in this article will attract the interest of [23] Creating HIPAA-Compliant Medical Data Applications with Amazon
Web Services, White Paper, available online at:
scientists, developers and industrial partners working in the http://awsmedia.s3.amazonaws.com/AWS_HIPAA_Whitepaper_Final.
field of biomedical informatics. This paper has presented pdf
@HealthCloud: a prototype implementation of a mobile
1040

View publication stats