Information Technology Reference
In-Depth Information
domains, such as healthcare or eHealth (Ranz,
Aud, Alexander, Oliver, Minner, & Skubic, 2008).
A majority of AmI projects follow a hardware-
oriented (bottom-up) development methodology.
This also corresponds with the findings in (Kur-
schl, Schoenboeck, & Mitsch, 2009) and (Buch-
mayr & Kurschl, 2010). A possible explanation
is that most projects prototypically implement the
use cases first and care about integration issues
later. In contrast, a model driven approach allows
creating a more interoperable system, although
there still remain open issues concerning the
integration of AAL and healthcare data. On the
one hand the issues are related to the different
purpose of these systems. While AAL systems
focus on perceiving and processing environmental
information, healthcare applications often focus
on processing document-based patient data. On
the other hand a lack of applied standards, like
HL7, and the question how to properly apply them
to the processed data within AAL systems cause
problems. Figure 4 shows the information gap
related to the hardware driven approach in contrast
to the information gap related to the application/
model driven approach. For creating an integrated
system, which includes technologies from the
domains AAL and healthcare, a model driven
development methodology should be favored.
Emergency situation treatment is one of the
core requirements for AAL systems. The follow-
ing section will be used to explain concepts for
describing situations based on models. Such
models are not only needed for situation descrip-
tion they also facilitate the transformation and
exchange of data with other related systems, like
healthcare systems. Furthermore the processing
of health data and the exchange of patient care
data will be explained by using state of the art
projects as reference.
Situation Detection and
Situation Awareness
The prerequisite for proper situation detection is,
to gain Situation Awareness . Most of the situation
awareness methodologies described in this chapter
are based on the research of Endsley (Endsley M.,
1998), (Endsley M. R., 1995) and their applica-
tion by the data fusion community (Steinberg,
Bowman, & White, 1998). Endsley analyzed the
behavior of human beings and defined Situation
Awareness as: “… the perception of elements
in the environment within a volume of time and
space, the comprehension of their meaning and
the projection of their status in the near future
…” (Endsley M., 1998). In a nutshell, Situation
Awareness means to become aware of the current
situation and to conclude about a possible future
development.
According to (Llinas, Bowman, Rogova, &
Steinberg, 2004) and (Steinberg, Bowman, &
White, 1998) the JDL Data Fusion Model defines
a partitioning scheme, which consists of four dif-
ferent levels as shown in Table 1). The process
Table 1. Data Fusion partitioning scheme according to (Llinas, Bowman, Rogova, & Steinberg, 2004)
Data Fusion Level
Association Process
Estimation Process
Product
L0 - Signal Assessment
Assignment
(Observation-to-Feature)
Detection
Estimated Signal State
L1 - Object Assessment
Assignment
(Observation-to-Entity)
Attributive State
Estimated Entity State
L2 - Situation Assessment
Relationship
(Entity-to Entity)
Relation
Estimated Situation State
L3 - Impact Assessment
Evaluation
(Situation-to-Actor's Goals)
Game Theoretic
Interaction
Estimated Situation Utility
Search WWH ::




Custom Search