Segmentation and Interpretation of MR Brain Images Using an Improved Knowledge-Based Active Shape Model

An improvement of the Active Shape procedure introduced by Cootes and Taylor is presented. The new automated brain segmenta-tion and interpretation approach incorporates a priori knowledge about neuroanatomic structures and their speciic structural relationships to provide robust segmentation and labeling. The method was trained in 8 MR brain images and tested in 19 brain images by comparison to observer-deened independent standards. Neu-roanatomic structures in all images from the test set were successfully identiied. The presented method is applicable to virtually any task involving deformable shape analysis.