Image, Signal, and Distributed

PATIENT INFORMATION—ANYTIME ANYWHERE

Data Processing for Networked

eHealth Applications
A View from the Guest Editors
© ARTVILLE, LLC

ILIAS MAGLOGIANNIS, MANOLIS WALLACE,

AND KOSTAS KARPOUZIS

C
omputer-based patient record systems are con- knowledge, and make decisions in a given context [9].
tinuously expanding in order to support more Natural user interactions with such applications are based
clinical activities and serve healthcare profes- on autonomy, avoiding the need for the user to control
sionals more efficiently. For this reason health- every action, and adaptivity, so that they are contextual-
care organizations and providers are asking their staff to ized and personalized, delivering to the medical person-
interact more often with computer systems during their nel the right information and decision at the right
everyday work. Hospitals and healthcare centers are moment [10].
nowadays sufficiently rich in their computing infrastruc- A typical application falling into the specific concept,
ture to handle the internal administrative and clinical for instance, is pervasive patient telemonitoring (an
processes for both inpatients and outpatients. The data important part of telemedicine), which involves the sens-
stored in computer-based patient record systems include ing of a patient’s physiological and physical parameters
medical documents relating to the past, present, or future and transmitting them to a remote location, typically a
physical condition of a patient, the results of examina- medical center, where expert medical knowledge resides.
tions in the form of multimedia (text, medical images, A typical telemonitoring system has the ability to record
sounds, and videos), and financial and demographic physiological parameters and provide information to the
information [1], [2]. doctor in real time through a wireless connection, while it
In addition, the rapid development in communications requires sensors to measure parameters like arterial blood
through fixed or mobile networks has opened new ways in pressure, heart rate, electrocardiogram, skin temperature
computer-based health systems by providing the capabili- and respiration, glucose, or patient position and activity.
ty of remote and distributed access to patient’s medical Filtered signals and medical data are either stored locally
data [3], [4]. Remote patient monitoring in terms of on a monitoring wearable device for later transmission or
telemedicine and the provision of clinical guidelines used directly transmitted, e.g., over the public phone network,
for the patient's care from distant locations [5] are sup- to a medical center [11]–[13].
ported within networked computer-based patient record Thus, the field of networked eHealth has clearly already
systems, while retrieval of medical data [6] and remote shown its potential, facilitating exchange of information
teleconsultations between healthcare professionals are between clinicians or between institutions, reducing costs,
also possible [7]. Medical data are captured and transmit- extending the scope and reach of medical facilities,
ted, received or updated, stored or retrieved securely and enhancing the quality of service offered on- and off-site,
in real time by users in geographically distributed and providing new means of medical supervision and preemp-
organizationally independent organizations or distant tive medicine, and so on. Currently, the integration of
locations. medical networking and medical information systems is
In this era of distributed computing the trend in med- treated as an obvious and irrefragable rule. Although
ical informatics is toward achieving two goals: the avail- major progress has been made with respect to the integra-
ability of software applications and medical information tion of distributed processing and communication capabil-
anywhere and anytime and the invisibility of computing; ities offered, the field is still considered to be at a
computing modules are hidden in multimedia information premature stage. In addition to interchangeability and data
appliances that are used in everyday life [8]. Both afore- exchange, computer science can also offer intelligent
mentioned goals require distributed data processing mod- computing services to the medical sector; thus, clinical
ules that will be able to automatically analyze data decision support is a main focus. In this framework, state-
provided by medical devices and sensors, exchange of-the-art image, signal, and data processing is required,
so that useful information is extracted and reliable auto-
Digital Object Identifier 10.1109/EMB.2007.901781 mated recommendations are achieved.

14 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE 0739-5175/07/$25.00©2007IEEE SEPTEMBER/OCTOBER 2007

 In this era of distributed computing the trend in
medical informatics is toward achieving two goals:
the availability of software applications and
medical information anywhere and anytime
and the invisibility of computing.

In this Issue eHealth. Although the results of this progress are certainly
Within this special issue we review some of the most beneficial and promising, it is becoming obvious that
promising prospects for the state-of-the-art and emerging some important technological factors may have been over-
field of image, signal, and distributed data processing for looked in the initial enthusiasm. Specifically, we may
networked eHealth applications. have originally underestimated the way in which the sud-
The case of pressure ulcer detection is an appropriate den availability of cheap and trustworthy storage and
example for this. With the number of individuals that communication solutions would affect the amount of med-
belong to high-risk social groups for the development of ical data retained in digital format.
pressure ulcers, it is quite difficult for a national health- More and more medical data are becoming available in
care system to have sufficient expert medical personnel digital formats and to greater details. Although this pro-
distributed throughout a country and dedicated to the vides more processable information and thus enhanced
detection and treatment of pressure ulcers. Kosmopoulos computing opportunities, it also augments the require-
and Tzevelekou [19] go further than existing manual tech- ments in storage and bandwidth, often overweighting the
niques and propose an automated system for the analysis rise in available resources. Clearly, we desire to have
of two-dimensional (2-D) images and classification of the techniques that will limit the size of the acquired data
stage of pressure ulcers present in them. Following a sup- but not the information richness, thus enhancing the bal-
port vector machine approach, they achieve high rates of ance between available and required resources for opti-
tissue classification to healthy or one of six different mized medical services. Doukas and Maglogiannis in
stages of pressure ulcers. Such a system can be utilized as [15] provide an excellent review of such techniques in
a first filtering step, so that the amount of data that has to the field of medical image compression, the medical data
be manually processed by clinicians is both reduced and that consumes the vast majority of available resources.
labeled for gravity. Region of interest coding is analyzed and compared and
In the development of such systems, the availability of classified in 2-D, three-dimensional (3-D), and video
training data is often an issue. Clinical decision support sequence categories.
is most desired in rare medical cases for which a single In all of the above we have assumed that distributed
clinician usually cannot have adequate practical experi- medical repositories cooperate fully and, thus, that the
ence. In such cases, analyzing medical history records only considerations for data exchange are related to data
from a single clinic or medical facility does not typically format, channel capacity, and security. Unfortunately, this
suffice for the development of a robust and reliable high- also constitutes an oversimplification. In reality, in most
accuracy system. Thus, we need to combine data residing cases individual medical archives are bound by legal con-
at different locations. straints not to freely redistribute the medical history
Additionally, to the ordinary data format and channel records they possess; ownership and cost issues may also
bandwidth considerations that govern most data intercon- interfere. As a result, owners of medical information are
nection problems, in the medical case we also need to often quite reluctant to permit its transmission to other
consider legal and ethical issues related to data ownership sites so that it can be processed remotely.
and confidentiality. Medical data transfer is also required One of the main fields in which electronic healthcare
in order to offer medical services at the time and place has been making a major contribution to clinical medicine
they are needed, by making centralized services available is clinical decision support. In this framework, medical
electronically through permanent and ad-hoc networks. In history data is processed automatically using state-of-the-
both cases, healthcare information needs to be accessible art algorithms in order to develop automated recommen-
by authorized users only, while its fundamental security dation systems. With today’s augmented computing
properties, namely integrity, availability, and confiden- power resources, using the best available processing algo-
tiality, must be retained. The technological challenges for rithms is not an issue. What is an issue, on the other hand,
this are presented and discussed by Gritzalis et al. in [18]. is the availability of medical history data due to the fact
Vastly augmenting communication bandwidths and that issues related to security, ownership, and confidential
rapidly diminishing costs for digital storage have allowed personal medical data prohibit the gathering of this data to
the realization of the above-mentioned integrated systems a main medical database.
that handle or exchange large amounts of medical data. Given the statistical nature of utilized algorithms, the
Following this trend, the acquisition, digitization, storage, independent processing of the data and the combination of
remote access to, and analysis of large amounts of med- the individual results cannot provide overall results of
ical data are becoming typical things in the field of equivalent quality to those of centralized processing of all

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE SEPTEMBER/OCTOBER 2007 15

 Although major progress has been made with
respect to the integration of distributed processing
and communication capabilities offered, the field
is still considered to be at a premature stage.

the data. Megalooikonomou and Kontos in [17] propose a Unfortunately, the amount and diversity of high-risk
novel hierarchical processing algorithm in order to tackle groups and individuals makes this, too, an unrealistic
this. In their approach, although data are only processed at option. Additionally, few people would be content with
their original location so that confidentiality issues do not this extent of intervention in their day-to-day life.
arise, the distributed processing components are in con- Pervasive, automated, citizen-centered care is probably
stant communication with each other throughout the pro- the most important emerging technological achievement
cedure, thus achieving a coordination that allows for in the scope of distributed eHealth. In this framework,
results equivalent to those acquired when applying cen- automated digital systems monitor an individual belong-
tralized processing algorithms. ing to a high-risk group, providing both personalized
With methodologies such as the above we can generate lifestyle-related advice and early diagnosis in the individ-
automated systems able to process medical information ual’s own environment with little or no interference with
and provide decision support recommendations to clini- everyday life. This kind of personalized care was not
cians; this might not be the only field of application for available in the past outside the clinic and without the
distributed eHealth. It is becoming more and more obvious support of clinicians and specialized technicians due to
that healthcare systems cannot possibly offer full coverage the cost, size, and complexity of the required equipment.
for the geographic and population spread of a country. On Recent advances in electronics have made possible the
the other hand, we now know that asymptomatic diseases production of miniscule medical observation and process-
are difficult to detect at an early stage without comprehen- ing devices that come in many flavors, with portable and
sive medical examinations; without an early detection the wearable ones being the most promising. Out of these, the
chances for treatment are severely narrowed. The only wearable type, being the noninvasive one, is the one
realistic solution is to identify high-risk population groups receiving most attention and providing the best prospects
so that medical advice and monitoring can be targeted. for the future. Gatzoulis and Iakovidis in [16] provide a
In [14] Kyriacou et al. see the utilization of automated rich presentation of the state of the art, emerging innova-
analysis techniques in the field of medical risk analysis. tions, and new opportunities in the field.
With stroke being the leading cause of death in Western
societies, and with the process building up to it often Acknowledgments
being free of symptoms, the identification of high-risk The guest editors wish to thank all the authors for their
individuals or groups is desired. Such information can contributions, all reviewers for their efforts and valuable
help greatly reduce strokes and related deaths or resulting reviews, and especially Editor-in-Chief John Enderle for
disabilities. In the work presented in this article, in order his advice and encouragement. Although the topics dis-
to acquire this type of knowledge, medical data have been cussed in this special issue could not certainly cover the
collected from a high number of patients using noninva- whole spectrum of distributed eHealth applications, we do
sive techniques; this data was then analyzed with novel hope that the articles included will attract the interest of
automated processing techniques in order to extract infor- this publication’s audience and serve as an inspiration to
mation-rich features. The patients were then monitored for the young researchers working in this field.
a period reaching up to 84 months. The study of the statis-
tical correlation between the data collected from patients Ilias Maglogiannis received a Diploma in
and their latter health status and progress is utilized in the electrical and computer engineering and a
developed medical portal in order to automatically pro- Ph.D. in biomedical engineering and med-
vide estimations of whether specific individuals or groups ical informatics from the National
belong to high-risk categories. With both population and Technical University of Athens (NTUA),
diversity of illnesses augmenting rapidly, it is not possible Greece, in 1996 and 2000, respectively,
to provide constant and detailed medical analysis for all with a scholarship from the Greek govern-
people and in all cases, so works such as this that utilize ment. From 1996 until 2000 he worked as
computerized technology to identify high-risk groups pro- a researcher in the Biomedical Engineering Laboratory at
vide a much more realistic approach to preemptive medi- NTUA, and he has been involved with several European and
cine and early detection. national projects. In 2001 he joined the faculty of the
Once someone has been identified as a member of a Department of Information and Communication Systems
high-risk group, we need to be able to provide personal- Engineering at the University of the Aegean. His published
ized advice and pervasive monitoring. Ideally, a personal scientific work includes two books and five lecture notes (in
or family clinician would undertake the task. Greek) on biomedical engineering and artificial intelligence

16 IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE SEPTEMBER/OCTOBER 2007

topics, 26 journal papers, and more than 40 international Address for Correspondence: I. Maglogiannis, Dept. of
conference papers. He has served on program and organiz- Information & Communication Systems Engineering,
ing committees of national and international conferences University of the Aegean, 83200 Karlovasi, Greece Tel:
and he is a reviewer for several scientific journals His scien- +30 22730 82239, Fax: +30 22730 82008. E-mail: ima-
tific activities include biomedical engineering, telemedicine glo@aegean.gr.
and medical informatics, image processing, and multimedia
telecommunications. He is a member of the IEEE References
Engineering in Medicine and Biology Society, IEEE [1] I. Maglogiannis, N. Apostolopoulos, P. Tsoukias, “Designing and imple-
Computer Society, IEEE Communications Society, the menting an electronic health record for personal digital assistants (PDAs),”
International Society for Optical Engineering, the ACM, the Int. J. Qual. Life Res., vol.2, no. 1, pp. 63–67, 2004.
Technical Chamber of Greece, the Greek Computer Society, [2] A. Dalley, J. Fulcher, D. Bomba, K. Lynch, and P. Feltham, “A technologi-
the Hellenic AI Society, and the Hellenic Organization of cal model to define access to electronic clinical records,” IEEE Trans. Inform.
Biomedical Engineering. Technol. Biomed., vol.9, no.2, pp. 289–290, 2005.
[3] E. Mendonça, E. Chen, P D. Stetson, L. McKnight, J. Lei, and J. Cimino,
Manolis Wallace was born in Athens in “Approach to mobile information and communication for health care,” Int. J.
1977. He obtained his Diploma in Electrical Med. Inform., vol. 73, no. 7-8, pp. 631–638, 2004.
and Computer Engineering from the [4] E. Hall, D.K. Vawdrey, C.D. Knutson, and J.K. Archibals, “Enabling
National Technical University of Athens remote access to personal electronic medical records,” IEEE Eng. Med. Biol.
(NTUA) and his Ph.D from the Computer Mag., vol. 22, no. 3, pp. 133–139 2003.
Science Division of NTUA. He is with the [5] C. Finch, “Mobile computing in healthcare,” Health Manage.Technol., vol.
University of Indianapolis, Athens Campus 20, no. 3, pp. 63–64, 1999.
(UIA) since 2001 where he serves now as [6] G. Kambourakis, I. Maglogiannis, and A. Rouskas, “PKI-based secure
an Assistant Professor. Since 2004 he is also the Chair of the mobile access to electronic health services and data,” Technol. Healthcare,
Department of Computer Science of UIA. vol. 13, no. 6, pp. 511–526, 2005.
His main research interests include handling of uncer- [7] I. Maglogiannis, K. Delakouridis, and L. Kazatzopoulos, “Enabling collab-
tainty, information systems, data mining, personalization orative medical diagnosis over the Internet via peer to peer distribution of
and applications of technology in education. He has pub- electronic health records,” J. Med. Syst., vol. 30, no. 2, pp. 107–116 2006.
lished more than 40 papers in the above fields, ten of [8] G. Abowd, “Software engineering issues for ubiquitous computing,” in
which in international journals. He is the guest editor of Proc. Int. Conf. Software Engineering, Los Angeles, 1999, pp. 5–84.
two journal special issues and a reviewer for more than [9] G.E. Barnes and S. Warren, “A wearable, Bluetooth-enabled system for
ten journals and various books. His academic volunteer- home health care,” in Proc. 2nd Joint EMBS/BMES Conf., Houston, TX, Oct.
ing work also includes participation in various confer- 2002, pp. 1879–1880.
ences as organizing committee chair, session organizer, [10] E. Jovanov, “Stress monitoring using a distributed wireless intelligent
session chair or program committee member. He is the sensor system,” IEEE Eng. Med. Biol. Mag., vol. 22, no. 3, pp. 49–55, 2003.
general chair of SMAP 2006. [11] V. Rialle, J. Lamy, N. Noury, and L. Bajolle, “Telemonitoring of patients
at home: A software agent approach,” Comput. Methods Programs Biomed.,
Kostas Karpouzis is an associate re- vol. 72, no. 3, pp. 257–268, 2003.
searcher at the Institute of Communi- [12] K. Hung and Y. Zhang, “Implementation of a WAP-based telemedicine
cation and Computer Systems (ICCS) system for patient monitoring,” IEEE Trans. Inform. Technol. Biomed., vol. 7,
and holds an adjunct lecturer position at no. 2, pp. 101–107, 2003.
the University of Piraeus, teaching data [13] J. Andreasson et al., “Remote system for patient monitoring using
warehousing and data mining. He gradu- Bluetooth/spl trade” Sensors 2002, Proc. IEEE, vol. 1, pp. 304–307, June
ated from the School of Electrical and 2002.
Computer Engineering of the National [14] E. Kyriacou, C.S. Pattichis, M. Karaolis, C. Loizou, C. Christodoulou,
Technical University of Athens in 1998 and received his M.S. Pattichis, S. Kakkos, and A. Nicolaides, “An Integrated system for
Ph.D. degree in 2001 from the same University. His cur- assessing stroke risk,” IEEE Eng. Med. Biol. Mag., vol. 26, no. 5, pp. 43–50,
rent research interests lie in the areas of human computer 2007.
interaction, image and video processing, image inter- [15] C. Doukas and I. Maglogiannis, “Region of interest coding techniques for
change infrastructures using the MPEG-4 and MPEG-21 medical image compression,” IEEE Eng. Med. Biol. Mag., vol. 26, no. 5, pp.
standards, sign language synthesis, and virtual reality. 29–35, 2007.
Dr. Karpouzis has published more than 70 papers in [16] L. Gatzoulis and I. Iakovidis, “Wearable and portable eHealth systems,”
international journals and proceedings of international IEEE Eng. Med. Biol. Mag., vol. 26, no. 5, pp. 51–56, 2007.
conferences. He is a member of the technical committee [17] V. Megalooikonomou and D. Kontos, “Medical data fusion for telemedi-
of the International Conference on Image Processing cine: A model for distributed analysis of medical image data across clinical
(ICIP), the IFIP Conference on Artificial Intelligence information repositories,” IEEE Eng. Med. Biol. Mag., vol. 26, no. 5, pp.
Applications and Innovations (AIAI) and a reviewer in 36–42, 2007.
many international journals. Since 1995 he has partici- [18] S. Gritzalis, P. Belsis, and S.K. Katsikas, “Interconnecting autonomous
pated in more than 12 R&D projects at Greek and medical domains,” IEEE Eng. Med. Biol. Mag., vol. 26, no. 5, pp. 23–28,
European level. He is also a national representative in 2007.
IFIP Working Groups 12.5 “Artificial Intelligence [19] D. Kosmopoulos and F. Tzevelekou, “Automated pressure ulcer lesion
Applications” and 3.2 “Informatics and ICT in Higher diagnosis for telemedicine systems,” IEEE Eng. Med. Biol. Mag., vol. 26, no.
Education.” 5, pp. 18–22, 2007.

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE SEPTEMBER/OCTOBER 2007 17