Nowadays the availability of a huge amount of raw medicaldata makes it possible to use suitable data mining techniques to producenew knowledge. Usually, only data mining experts are able enough tocarry out such tasks, and not so many researchers in medical ¯eld arealso skilled in data analysis. This paper describes the Medical KnowledgeDiscovery Assistant (MKDA), a web based framework able to advice amedical researcher in such tasks. MKDA plans a Knowledge DiscoveryProcess (KDP) on the basis of the requests of the user and of a set ofrules in a knowledge base. The requests of the user are related to accu-racy, computational load, type of the produced model. They de¯ne thegoal to be reached by the planned process. The whole system relies ona database that contains medical data, e.g. images, text, examinationresults. The rules consider the logic schema of the database that is de-scribed semantically. The system's work is the result of the co-operationof di®erent web services specialized in di®erent tasks. The web servicesare chosen on the basis of their functionalities described by a commonontology. This ontology allows to integrate the process information anddomain information about the knowledge discovery process. The paperpresents a possible scenario and an experiment dealing with region clas-si¯cation of medical images.
|Numero di pagine||11|
|Stato di pubblicazione||Published - 2008|
All Science Journal Classification (ASJC) codes