Medical knowledge extraction. Applications of data analysis methods

• In this thesis we explore and discuss some important methods for knowledge extraction from meclical data. This is done in relation to, and for the purpose of design and development of decision support systems, which could be population specific.To test data and extract knowledge, we use univariate and multivariate statistical methods, the rough sets theory and probabilistic artificial intelligence approaches. These methods are used to estimate characteristics of patient groups, disease profiles and other features relevant for medical problems. In particular, we apply them to clifferentiate among patient groups, develop patient models and derive decision rules. Our experience refers to two medical domains (patients with diagnosed and non-diagnosed, but suspected liver disease and patients with duodenal ulcer surgery).Extracted knowledge can be used both in clinical practice and health care programs, as well as in computer based decision support systems to adjust them to various clinical environments.

Nyckelord

MEDICINE

MEDICIN

Publikations- och innehållstyp

vet (ämneskategori)

lic (ämneskategori)