SEGMENTATION OF MRI VOLUME DATA BASED ON GEOMETRIC ACTIVE CONTOUR AND MORPHOLOGICAL CHARACTERISTICS

We analyze the difficulties of segmentation in ventricle MR images, and propose an algorithm for segmentation of MRI volume data (VD) target profiles. So we propose Geometric active contour (GAC) morphological characteristics model based on Hough operator and clustering method (MCS algorithm). Initially, using Hough transformation to automatically locate initial contour of VD, the algorithm uses clustering approach to complete data subsampling and initial segmentation. Next, according to the clustering rules, the proposed MCS algorithm finishes MRI VD segmentation. Finally, the algorithm uses a category optimization method to improve segmentation results. Experiments show that the algorithm provides good segmentation results.