Delineation of the pelvic girdle in computed tomographic images

Localization, identification, and segmentation of the thoracic, abdominal, and pelvic organs are important steps in computer-aided diagnosis, treatment planning, and content-based retrieval of biomedical images. In this context, to aid the landmarking and identification of the lower abdominal organs, to separate the abdominal cavity from the lower pelvic region, and to improve the process of localization of abdominal pathology, we propose methods to identify automatically the pelvic girdle. The methods include procedures to represent the pelvic surface by a quadratic model using linear least-squares estimation, and to refine the model using deformable contours. We have modified our previously proposed method to identify the spinal canal and used the related information to generate automatically seed voxels in the pelvic girdle. The opening-by-reconstruction procedure was used for segmentation of the pelvic girdle. The proposed procedure led to a reduction of 10.3% in the false-positive error rate in the segmentation of abdominal tumors caused by neuroblastoma.