Segmentation methods of CT cardiac images play an important role in quantitative cardiac image analysis methods and cardiac diagnostics. Methods that include extraction of statistical texture features for cardiac tissue classification are one of the most frequently used methods. Based on these methods, it can be automatically decided in which level of rejection the transplanted heart belongs. These methods are based on arrangement and spatial distribution of different gray level intensity pixels in an image. They are divided in two groups, first order statistical methods where histogram is included, and second order statistical methods where Gray Level Co-occurrence Matrix (GLCM) and Gray Level Run Length Matrix (GLRLM) are included. Every of these methods has its own statistical features which characterize image textures. Histogram is the simplest method because it doesn't consider spatial distribution of two neighbour pixels, instead, it takes pixel by pixel from the image and based on its gray level intensity iz calculates the features. Furthermore, GLCM is much more complicated method because it requires calculation of the matrices in four directions under four different angles (0°, 45°, 90° and 135°). The most famous statistical features of this method are Haralick features established by Robert Haralick et.al. in 1973. GLRLM is a statistical method based on calculation of run lengths in an image. As in GLCM, in this method are firstly calculated the matrices in four directions, and after that the characteristic features are calcualted.