Oral teeth image segmentation plays an important role in teeth orthodontic surgery and implant surgery. As the tooth roots are often surrounded by the alveolar, the molar’s structure is complex and the inner pulp chamber usually exists in tooth, it is easy to over-segment or lead to inner edges in teeth segmentation process. In order to further improve the segmentation accuracy, a segmentation algorithm based on local Gaussian distribution fitting and edge detection is proposed to solve the above problems. This algorithm combines the local pixels’ variance and mean values, which improves the algorithm’s robustness by incorporating the gradient information. In the experiment, the root is segmented precisely in cone beam computed tomography (CBCT) teeth images. Segmentation results by the proposed algorithm are then compared with the classical algorithms’ results. The comparison results show that the proposed method can distinguish the root and alveolar around the root. In addition, the split molars can be segmented accurately and there are no inner contours around the pulp chamber.
Objective To observe the difference between blood brain barrier and blood optic nerve barrier. Methods Twenty normal male SD rat sprime; optic nerve including prelaminar region, lamina cribrosa, retro-laminar region, intraorbital portion, intracanalicular portion, and intracranial portion respectively,and cerebral cortex were removed separately. Ultrastructure of endothelial cells was observed by electron microscopy. Immunohistochemical staining was used to detect the expression of transferrin receptor (OX-26) and metalloproteinase inducer (OX-47) and extravasation of fibrinogen around microvessels. Results The results of electron microscopy showed that endothelial cells of microvessels in each portion of optic nerves and cerebral cortex did share the same tight junctions. However, the number of plasmalemmal vesicles in prelaminar region was significantly more than that in cerebral cortex(P＜0.05);there was no significant difference between other parts of optic nerves (lamina cribrosa, retro- laminar region, intraorbital portion, intracanalicular portion, and intracranial portion)and cerebral cortex in the number of the plasmalemmal vesicles(Pgt;0.05). By immunohistochemical staining，the endothelial cells of microvessels in the prelaminar region showed no expression of the OX-26 and OX-47,but extravasation of fibrinogen around microvessels was found; b positive expression of OX-26 and OX-47 was observed in the endothelial cells of the microvessels in other parts of optic nerves (lamina cribrosa, retro-laminar region, intraorbital portion, intracanalicular portion, and intracranial portion) andcerebral cortex, and no fibrinogen was seen aro und the microvessels. Conclusions There is a significant difference between the endothelial cells of the microvessels in prelaminar region and cerebral cortex in the ultrastructure, markers expression, and permeability, so the microvessel s in prelaminar region lacks the typical blood brain barrier characteristics.The microvessels in other parts of optic nerves (lamina cribrosa, retro-laminar region, intraorbital portion, intracanalicular portion, and intracranial portion) have blood brain barrier properties due to its similar specialties as which in cerebral cortex. (Chin J Ocul Fundus Dis, 2006, 22: 390-393)
ObjectiveTo investigate the three-dimensional structure of proximal femoral trabeculae, analyze the formation mechanism, and explore its relationship with the occurrence and treatment of proximal femoral fractures.MethodsSix cadaver adult femur specimens were harvested and the gross specimens containing both trabecular system and cortical bone were established by hand scraping. All samples were scanned by micro-CT and the CT images were input into Mimics18.0 software to establish the digital proximal femoral model containing trabecular structure. The spatial distribution of trabecular system was observed, and the relations between trabecular bone and the proximal femur surface and related anatomical landmarks were analyzed in digital models.ResultsThe gross specimen and digital models of trabecular system were successfully established. The trabecular system of proximal femur could be divided into two groups: the horizontal and vertical trabecular. The horizontal trabecular arose from the base of greater trochanter, gone along the direction of femoral neck, and terminated at the center of femoral head. The vertical trabecular began from the base of lesser trochanter and femoral calcar, gone radically upward, and reached the femoral head. The average distance of the horizontal trabecular to the greater trochanter was 22.66 mm (range, 17.3-26.8 mm). In the femoral head, the horizontal trabecula and the vertical trabecula were fused into a kind of sphere, and the distances from the horizontal trabecula to the surface of the femoral head vary in different sections. The average distance of trabecular ball to the femoral head surface was 6.88 mm (range, 6.3-7.2 mm) in sagittal plane, 6.32 mm (range, 5.8-7.6 mm) in coronal plane, and 6.30 mm (range, 5.6-6.3 mm) in cross section. The vertical and horizontal trabeculae intersect obliquely, and the average angle of horizontal trabecular and vertical one was 140.67° (range, 129-150°).ConclusionThe trabecular system exhibits a unique spatial configuration, which is the main internal support of proximal femur. Restoration of the integrity of trabecular structure is the important goal of proximal femoral fractures.