Information Technology Reference
In-Depth Information
CONCLUSION
mapping between the domain ontology and data
sources is a straightforward mechanism and eas-
ily configurable.
The data governance ontology was expressed
in terms of roles and privileges of access. Roles
and Privileges are expressed in relation to the do-
main ontology concepts opposed to the individual
instances of data elements. This enabled dynami-
cally enacting roles in accordance with service data
governance policies and allowed extensibility not
only in defining new roles but also variations in
same roles across services with no physical co-
relation to or dependency on individual instances
or changes of mappings.
The meta-registry was implemented using a
modified version of the jUDDI version 3.0. It
was extended to allow storing semantic meta-
data about individual data services beyond the
simple service registration data that supports. The
meta-data included, in addition to WSDL-type
information, descriptions of the content, used
codings, available clinical data types in relation
to the domain ontology, data changes and service
support information. It also included annotations
of individual domain ontology mappings and data
governance constraints. These are used by the
ePCRN-IIA to enable dynamic discovery of and
binding to individual data services. Although, it
may appear that the current prototype relies on
the assumption that meta-data is kept up-to-date
for successful dynamic binding, our experiments
have shown little change tendency in the core
semantic elements that may affect the system. In
fact, in one case, it was a substantial change in the
underlying EHR system itself for which the data
service needed to be re-configured, including the
mapping ontologies. However, to be exhaustive
more experiments are needed to examine the extent
of change on its semantic-awareness extensibility.
Equally, the architecture may benefit from includ-
ing standardised semantic service descriptions,
such as those based on WSDL-S [Akkiraju, R. et
al (2005)] or OWL-S [Martin, D. et al (2004)],
which need evaluating for the next version.
This chapter presented the interoperability chal-
lenges in eHealth systems and argued that they
face similar challenges to those in service-oriented
architecture except with focus on data-intensive
needs. We assert that using semantic interoperabil-
ity standards is not sufficiently scalable approach
for complex data-intensive closed domains such as
eHealth, and that a domain-based semantic driven
approach is a more suitable adaptive solution.
The chapter outlined an approach to dynamic
semantic interoperability based on extensible
semantically enriched problem models and do-
main ontologies. It described briefly the relevant
implementation of the approach as part of the
ePCRN-IIA prototype and presented some ini-
tial results reflecting on its advantages, potential
requirements and limitations.
ACKNOWLEDGMENT
This work has been funded in part by the NIH
and NIHR. We would like to thanks all members
of the ePCRN project for their input and valu-
able contributions to the ePCRN development
of this paper.
REFERENCES
Akkiraju, R., Farrell, J., Miller, J., Nagarajan, M.,
Schmidt, M.-T., Sheth, A., & Verma, K. (2005).
Web service semantics-WSDL-S (Technical Note,
Version 1.0). Retrieved from http://lsdis.cs.uga.
edu/library/download/WSDL-S-V1.html
Ankolekar, A., Burstein, M., Hobbs, J. R., Las-
sila, O., Martin, D., McDermott, D., & Sycara, K.
(2002, June). DAML-S: Web service description
for the Semantic Web . International Semantic Web
Conference, Italy.
Search WWH ::




Custom Search