Sie befinden Sich nicht im Netzwerk der Universität Paderborn. Der Zugriff auf elektronische Ressourcen ist gegebenenfalls nur via VPN oder Shibboleth (DFN-AAI) möglich. mehr Informationen...
Interestingness Analysis of Semantic Association Mining in Medical Images
Ist Teil von
Computer Networks and Intelligent Computing, p.1-10
Ort / Verlag
Berlin, Heidelberg: Springer Berlin Heidelberg
Quelle
Alma/SFX Local Collection
Beschreibungen/Notizen
Medical images provide rich information and are repositories of implicit knowledge, which can be mined effectively. Diagnosis rules, new and established ones can be revealed and verified from the patterns existing in a medical image. In this paper, the spatial patterns existing in medical images are mined to obtain association rules of the diagnosis rules’ category. The semantics of the association rules obtained are highly improved by introducing the concept of fuzziness into the spatial relationships existing between the anatomical structures. The utility of the association rules extracted is analyzed through the interestingness measures computed, and it is thereby concluded that the rules mined are highly relevant.