Hello all,
My thesis "Detailed clinical models and their relation with Electronic
Health Records" is now available online. Until the university
publishes it on the public repository, here is a link to it.
https://www.dropbox.com/s/rqmmh9b541u7wpr/TesisDiego_v2.pdf?dl=0
The slides are also available just in case anyone wants a quick summary
I also attach the abstract below
Best Regards
Diego Boscá
Abstract
Healthcare domain produces and consumes big quantities of people’s
health data. Although
data exchange is the norm rather than the exception, being able to
access to all patient data is
still far from achieved. Current developments such as personal health
records will introduce
even more data and complexity to the Electronic Health Records (EHR).
Achieving semantic
interoperability is one of the biggest challenges to overcome in order
to benefit from all the
information contained in the distributed EHR. This requires that the
semantics of the
information can be understood by all involved parties. It has been
established that three layers
are needed to achieve semantic interoperability: Reference models,
clinical models
(archetypes), and clinical terminologies.
As seen in the literature, information models (reference models and
clinical models) are
lacking methodologies and tools to improve EHR systems and to develop
new systems that can
be semantically interoperable. The purpose of this thesis is to
provide methodologies and
tools for advancing the use of archetypes in three different scenarios:
- Archetype definition over specifications with no dual model
architecture native
support. Any EHR architecture that directly or indirectly has the
notion of detailed
clinical models (such as HL7 CDA templates) can be potentially used as
a reference
model for archetype definition. This allows transforming single-model
architectures
(which contain only a reference model) into dual-model architectures (reference
model with archetypes). A set of methodologies and tools has been developed to
support the definition of archetypes from multiple reference models.
- Data transformation. A complete methodology and tools are proposed
to deal with the
transformation of legacy data into XML documents compliant with the
archetype and
the underlying reference model. If the reference model is a standard then the
transformation is a standardization process. The methodologies and
tools allow both
the transformation of legacy data and the transformation of data
between different
EHR standards.
- Automatic generation of implementation guides and reference materials from
archetypes. A methodology for the automatic generation of a set of reference
materials is provided. These materials are useful for the development
and use of EHR
systems. These reference materials include data validators, example instances,
implementation guides, human-readable formal rules, sample forms, mindmaps, etc.
These reference materials can be combined and organized in different
ways to adapt
to different types of users (clinical or information technology
staff). This way, users can
include the detailed clinical model in their organization workflow and
cooperate in the
model definition.
These methodologies and tools put clinical models as a key part of the
system. The set of presented methodologies and tools ease the
achievement of semantic interoperability by providing means for the
semantic description, normalization, and validation of existing and
new systems.
Does this mean you've passed the oral successfully? If so big congrats!!!