engleski

Multiresolution CT head image analysis using simulated annealing

Segmentation of computed tomography (CT) head images is an important step in quantitative analysis of human spontaneous intracerebral brain hemorrhage (ICH). A new multiresolution probabilistic approach for segmentation of CT head images containing ICH region is presented in this work. In the proposed method, the segmentation problem is viewed as a pixel labeling problem. In this particular application the labels are: background, skull, brain tissue, and ICH. The proposed method is based on the Maximum A-Posteriori (MAP) estimation of the unknown pixel labels (i.e. the segmented image). A Markov random field (MRF) model has been used for the posterior distribution. The MAP estimation of the segmented image has been determined using the simulated annealing (SA) algorithm. The multiresolution approach has been applied in order to speed up the SA algorithm. Experimental results have demonstrated good results and proved the usability of the method.

image segmentation; simulated annealing; multiresolution; computed tomography

nije evidentirano