Title :
MR brain imaging segmentation based on spatial Gaussian mixture model and Markov random field
Author :
Peng, Zhigang ; Wee, William ; Lee, Jing-Huei
Author_Institution :
Electr. adn Comput. Eng. & Comput. Sci., Cincinnati Univ., OH, USA
Abstract :
We present a novel method to effectively segment the three dimensional MR brain images (volumes) with severe intensity nonuniformity. The segmentation problem was formulated using maximum a posterior probability and Markov random filed (MAP-MRF) framework. A novel spatial Gaussian mixture model (SGMM) is used to represent the intensity probability distribution of each of the three brain tissues (WM, GM and CSF), and MRF is used to compute the prior probability. This method consists of a learning process based on expectation maximization algorithm (EM) to estimate the parameters of SGMM, and a classification algorithm based on iterated conditional modes (ICM) to perform the segmentation of the sequential brain images using the parameters obtained from the learning process. The results on the simulated and twenty in vivo MR brain volumes demonstrate the efficiency of this method. We also present the comparison results with other published methods.
Keywords :
Gaussian processes; Markov processes; biomedical MRI; brain; expectation-maximisation algorithm; image segmentation; MR brain imaging segmentation; Markov random field; brain tissues; expectation maximization algorithm; intensity nonuniformity; intensity probability distribution; iterated conditional modes; learning process; maximum a posterior probability; parameters estimation; sequential brain images; spatial Gaussian mixture model; Biomedical computing; Biomedical engineering; Biomedical imaging; Brain modeling; Coils; Image analysis; Image segmentation; Markov random fields; Pixel; Radio frequency; Gaussian mixture model; Markov random filed; image segmentation; maximum a posterior probability; medical image analysi;
Conference_Titel :
Image Processing, 2005. ICIP 2005. IEEE International Conference on
Print_ISBN :
0-7803-9134-9
DOI :
10.1109/ICIP.2005.1529750