Information Technology Reference
In-Depth Information
13.1
Introduction
Medical data stored in clinical files and databases, such as patient histories and
medical records, as well as research data collected for various clinical studies, are
invaluable sources of medical knowledge. The rapidly increasing number of elec-
tronically stored patients' data provides a tremendous opportunity for data mining,
the process of automatically discovering useful information (discovering knowl-
edge) in large data repositories [4]. Data-mining techniques allow for discovering
patterns, relationships, trends, typical cases, and irregularities in these large vol-
umes of data. These newly discovered information and knowledge can be used to
stimulate further research, as well as to create practical guidelines for diagnosis,
prognosis, and treatment. Thus, a successful data-mining process may result in a
significant improvement in the quality and efficiency of both medical research and
health care services. Many studies have already demonstrated the practical values
of data mining in various fields. However, in contrast with more traditional areas of
data mining, such as mining of financial data or mining of purchasing records, med-
ical data-mining presents greater challenges. These challenges arise not only from
the complexity of the medical data, but more fundamentally from the difficulty of
defining medical concepts. Medical concepts must be clearly defined in order to
build appropriate data models in the data-mining process. Thus, although comput-
ers allow us to store and process increasingly large volumes of data, the problem lies
in the creation of the suitable conceptual models for the data. These models should
be unambiguously defined, and they should be explicitly connected with the related
medical concepts. Evidently, the quality of the data-mining process depends on the
quality of the conceptual data models and the quality of the data.
This paper addresses issues specific to conceptual modeling of medical data in
the data-mining process. We will situate our discussion in the context of Dr. Kazem
Sadegh-Zadeh's typology of medical concepts [17]. In his
Handbook of Analytic
Philosophy of Medicine
, Dr. Sadegh-Zadeh outlines four main classes of medical
concepts: individual, qualitative (classificatory), comparative, and quantitative. Fur-
thermore, Dr. Sadegh-Zadeh divides medical concepts into classical concepts and
non-classical concepts. We demonstrate how Dr. Sadegh-Zadeh's typology can be
utilized for conceptual modeling of medical data. Specifically we illustrate how this
typology pertains to concepts and data used in the diagnosis and treatment of sleep
disorders. The paper is structured as follows. Section 13.2 defines the key issues
in modeling of medical concepts: definition of a concept, characteristics of medical
concepts, and computational representation. Section 13.3 describes the fundamen-
tal issues in medical data mining and provides the example of operational definition
for obstructive sleep apnea.
Section 13.4 presents the semiotic approach to con-
ceptual modeling.
The final section, Section 13.5, provides the conclusions and
future work.
