Information Technology Reference
In-Depth Information
INTRODUCTION
The integration of data from various sources
is a typical
data fusion
problem. One possible
solution is to apply the
JDL Data Fusion Model
(Llinas, Bowman, Rogova, & Steinberg, 2004).
The process described by this model covers the
perception of sensor data, the detection of specific
situations and the evaluation of a possible situation
evolution. Besides, the JDL Data Fusion Model
there are other models and/or processes that cope
with the problem of multi sensor data fusion,
like Luo and Kay's architecture for information
systems (Paradis & Treurniet, 1998), Pau's Sen-
sor Data Fusion Process (Esteban, Starr, Willetts,
Hannah, & Bryanston-Cross, 2005), the Omnibus
Data Fusion Model (Bedworth & O'Brien, 2000)
or the Waterfall Data Fusion Model (Bailey, Dodd,
& Harris, 1998). Most are based on the findings
of Endsley, who developed a model of situation
awareness in dynamic decision making based on
human beings (Endsley M. R., 2000), (Endsley
M., 1998).
A situation aware system, based on data fusion,
can support eldercare in the field of homecare,
but lacks the ability to integrate data from dif-
ferent external data providers (hospitals, care
institutions) in a standardized way. To solve
such integration issues researchers, like (Kokar,
Matheus, & Baclawski, 2009) and (Baumgartner
& Retschitzegger, 2006) propose models for data
fusion and situation awareness based on ontologies
to support knowledge sharing among different do-
mains. Transforming theses concepts into the AAL
domain enables electronic exchange of patient
data and establishes compliance with international
standard recommendations, like
Integrating the
Healthcare Enterprise
(IHE).
Efficient exchange of electronic patient record
data between different health service environments
is a challenge, since:
(ii)
semantically correct ex-
change of patient data requires appropriate care
data models and
(ii)
incomplete or contradictory
care information must be identified and completed
or corrected (intelligent mapping process). Stan-
Due to population aging and longer life expectancy
our society faces serious challenges. This concerns
especially the growing group of elderly people
(Madsen, Serup-Hansen, & Kristiansen, 2002).
One important challenge for elderly people is to
extend the time they can remain (independently)
in their familiar environment (Rennemark & al.,
2009). Pervasive home healthcare services and
systems help to maintain the health and func-
tional capabilities of individuals, support people
with chronic illnesses or disabilities and are one
prerequisite for an independent ageing at home.
Nowadays, pervasive home healthcare systems
are often composed of numerous independent
liv-
ing assistants
with the purpose to support elderly
persons and persons with special needs in their
activities of daily living. Proposed living assis-
tants often include features, like item tracking and
searching, warning of household dangers (slippery
floor, unattended stove, running water taps), recog-
nition of critical situations (collapse, fall), as well
as remote health monitoring and data exchange
with nursing facilities and care institutions. The
coherent information gained from various living
assistants increases the scope of detectable situa-
tions, improves the fault tolerance of the systems,
and leads to a richer user experience.
Various projects within the
Ambient Assisted
Living
(AAL) domain have achieved remarkable
results by using wireless sensor technology for
data collection, but still face problems concerning
the exchange and integration of healthcare data
(Wozak, Ammenwerth, Hörbst, Sögner, Mair, &
Schabetsberger, 2008), (Burgsteiner & al., 2009).
Thus, feasible healthcare systems have to cope
with several problems:
(ii)
patient data exchange
across institutional borders, such as nursing homes,
hospitals or AAL equipped retirement homes,
(ii)
data integration from various sources, and
(iii)
automated processing of unstructured or semi
structured data.
Search WWH ::
Custom Search