Information Technology Reference
In-Depth Information
A scheme summarizing of the process is shown in Figure 23.8.
After every access to the system, the dynamic pertinence is automatically updated
so the next time a context is acceded the pertinence of the data groups to the context
is computed considering the most updated information. As can be seen it is done
with a very low computation cost. In addition this updating process allows, not only
to give the most updated information, but also to discover new patterns of accesses,
which give us the chance to define new contexts.
23.5
Interoperability by Archetype-Based Contexts
Once we have presented the context-based access we will present the proposal for
interoperability. The proposals in this field are based on the interoperability be-
tween systems but forgetting the real users of the system. We want to change the
approach to give more importance to the users than the systems but allowing these
to communicate.
The main idea is to apply the context-bases access to the communication between
systems. In this case, we have to adapt the proposal to the especial situation where
the user and the information are in different hospitals or institutions. The contexts
on both hospitals are almost the same due to the medical praxis is very similar in
most of the institutions. We need to adapt the way to identify the context and the
representation for the information transmitted.
In the first case, in most of the country some information about the medical staff
that accesses the information has to be stored due to law obligation. Using this
information we can try to identify the context of the access. In this point we have to
differentiate three situations:
•
If this access is the first time that both systems interact, then it is complicated to
identify the context due to the normal differences between institutions. In this
case (Non-trained system) the system will give as answer a list of context related
to the medical staff specialty.
•
If there are some interactions in the past, the system can infer the context but an
error may occur so the system will answer the most probable context and a short
list of less probable context according to the past history (System under test).
•
If both systems have interacted previously in the past, enough to be be almost
sure about the context, the system will answer just the most probable context
(Trained system).
To implement this solution we only need three aspects: stored the previous accesses
(normally obligated by law), an API between the system (very simple) and a way
to send the archetypes. This is the last problem to solve: how to send the informa-
tion. As in the previous case, the solution depends on the systems that interact: if
both systems support an archetype language, then the information can be sent using
this language. In other case, a standard format for medical documents can be used
(DICOM [27] for images and PDF or XML for textual information).
With this points, all the aspects to adapt the solution are presented.
