Automated segmentation of the hip bones and cartilages in multi-slice CT images

摘要

In the hip joint degenerative diseases, quantification and visualization of the hip bone cartilages is of particular clinical interest. In this research, the datasets are acquired by injecting the contrast media in the hip joint during data acquisition under continuous leg traction. We present two independent methods for segmentation of the hip bones and cartilages. We first enhance the dynamic range and contrast of the data set by a conventional procedure. We then estimate the center of the femoral head utilizing a Hough transform. The estimated center is used as a pivotal point for beginning the segmentation process. Next, for cartilages segmentation, based on anatomical knowledge about the femoral head shape and size, we derive a region of interest (ROI) for further operations. In this case, by employing an Image Bottom Hat technique and anatomical knowledge about the curved shaped of the cartilages, we extract the valleys between contrast media and hip bones. For segmentation of bones, ROI are selected based on bones intensity. Then we utilize radial difference from femoral head center and Image Bottom Hat techniques for extracting bones edge in sagital view. The proposed model is validated on twenty sets (5120 images) of actual in vivo hip CT data.