Model-Based Segmentation of Femoral Head and Acetabulum from CT Images

Segmentation of bone tissue is an important topic in computer assisted medical image analysis. In particular, automatic segmentation of femoral head and acetabulum from CT dataset of hip joint respectively is of importance for various corresponding clinical tasks including visualization, quantitative assessment, early diagnosis and adequate management of patients suffering from avascular necrosis of the femoral head (ANFH). The non-uniformity of bone tissue, ranging from dense cortical bone to loose trabecular bone, the weakly defined joint space between femoral head and acetabulum and the inherent blurring of CT imaging render the segmentation of hip joint bones a challenging task. Traditional segmentation methods, e. g. thresholding, region growing and morphological can not well satisfy the segmentation demands. We then propose a statistical shape model based segmentation for bones in hip joint. In this paper, an improved algorithm of automatic landmarks selection for the generation of PDM for ASM is presented, the procedure of constructing ASM is described and the method using ASM to classify corresponding bone is analyzed and implemented.