Agent Technology and Healthcare: Possibilities, Challenges and Examples of Application - Computational Intelligence in Healthcare: Advanced Methodologies - page 32

Fig. 3. Communication-based scheduling of resources
using the facilities provided by the broker agent). The final ranking and filtering of
the set of alternatives received, takes into account the preferences of the user, which
selects one of them. The final approval of this alternative and rejection of the rest is
also made during this process.
In the case of enquiries , the treatment may require some missing values about the
patient. Usually, these values are findings, symptoms and syndromes that are used in
further steps of the treatment ( e.g., if a patient smokes, evaluation of a risk factor for
some disease). Also, like in the case of actions, several entries should be referred to
specialists. For instance, if an enquiry requires the result of a biopsy (positive or nega-
tive), this value should be filled by a surgeon who is able to perform it.
The last item that a practitioner can find in a CG is a decision that embeds logical
conditions and different paths that can be followed in the future. Decisions are made
by practitioners. HeCaSe2 only presents the information to them and waits for their
selection (in a supervised fashion). Being a decision support system, HeCaSe2 for-
wards all the information and data to the expert who decides upon all available possi-
bilities. In this case, it is not necessary to enable inter-communication processes
between agents.
4.3 Discussion
HeCaSe2 is a complex agent-based system that simulates the real behaviour of a
medical centre. HeCaSe2 guides the execution of daily care activities through the care
flow structure defined in CGs. This approach has two main benefits: first, the inclu-
sion of CGs in computerised health information systems, and second, to model a
flexible and robust platform with entities running autonomously but cooperating with
other partners to manage a more general problem [29].