Searching for new patterns in cardiovascular data

We study a new approach to rule induction from medical data sets with the help of physicians' assessment of the progressing solution in order to find new patterns in the available data. A method for the automatic extraction of rules is presented which is based on the evolutionary induction of decision trees and on automatic programming. The method is applied to a cardiovascular database. Several sets of rules are induced upon different groups of attributes which may possibly reveal the presence of some specific cardiovascular problems in young patients. The obtained rules are assessed by a physician in order to evaluate the strength of the developed knowledge discovery method.