Skip to Main Content
The expectation maximization (EM) algorithm has been used widely for computing the maximum likelihood (ML) parameters in the statistical segmentation of brain magnetic resonance (MR) images. As the standard EM algorithm is time and computer memory consuming, the segmentation is impractical in many real-world situations. In order to overcome this, an improved EM algorithm is presented. A novel statistical method is developed by combining the improved EM algorithm with a region growing algorithm, which is used to provide the a priori knowledge for the segmentation. The experimental results show that the proposed method can largely reduce the computing time and computer memory.