Clinical Semantics - Computational Intelligence in Healthcare: Advanced Methodologies - page 6

Chapter 2
Clinical Semantics
Jari Yli-Hietanen and Samuli Niiranen
Department of Signal Processing, Tampere University of Technology,
POB 553, 33101, Tampere, Finland
{jari.yli-hietanen,samuli.niiranen}@tut.fi
Abstract. We discuss the challenge of efficient and flexible clinical informatics
and provide initial results on how to tackle the computerized management of
the complex and diverse information space of clinical medicine through an
approach coined as open information management. The approach builds on
natural language as the core information management tool in place of formal,
structured representation. The chapter discusses a flexible, evolving clinical
practice supported by open clinical agents for both clinical professionals and
patients capable of learning at the human abstraction level. Clinical semantics is
not an add-on but rather natively integrated to, and an operational principle
behind, the functionality of these agents.
1 Introduction
1.1 Background
Figure 1 presents a framework for health care and life sciences with three key areas of
contemporary research activity: personalized medicine, translational medicine and
biosurveillance. Biomedical informatics is a well-recognized venue of research within
the health care, life sciences and information technology communities and is rightly
argued to be a key component in supporting advances in the three major areas of re-
search activity. It is also commonly accepted that the domain is a challenging one
from an informatics point-of-view with many open research problems.
Concerning the current status of informatics research in the domain, the application
of Semantic Web and related formal, structured approaches has been suggested to be a
key component in solving some of the challenges related to the complexity, diversity
and fluidity of information inherent to the domain. The thesis is that the use of these
technologies for information modeling, sharing and analysis will bring along a way to
approach the question of semantics in the domain in a unified and encompassing way.
However, a major practical argument can be made against the Semantic Web ap-
proach in bringing computable semantics to health care and life sciences domain
information. The argument emerges from the fact that, as an example of weak artifi-
cial intelligence, the approach primarily relies on the use of hand-crafted, metadata-
based annotations to provide for the semantics of information. In clinical informatics,
the approach relies on a massive knowledge and systems engineering effort to manu-
ally build integrating ontologies bridging the gaps between biological and clinical