Analysis of cardiac imaging data using decision tree based parallel genetic programming

We propose an algorithm for generating diagnostic rules for cardiac diagnoses. Diagnostic rules are presented in decision tree forms that are created by genetic programming. The algorithm was tested by using cardiac single proton emission computed tomography images. In comparisons with other six well-known methods including support vector machine, LogitBoost, logistic regression, linear discriminant analysis, linear regression and least square methods; the proposed algorithm is superior. We also show that parallel genetic programming can be used to improve the performance of the proposed algorithm.