Realistic virtual models can show periodontal and alveolar rich defects in three dimensions. Therefore, they can aid the surgical treatment process and provide a deeper understanding on postoperative healing mechanisms. Compared to previous and existing methods, the current approach displays each anatomical structure independently.
Therefore, these 3D virtual models represent the real clinical scenario realistically. This method helps to overcome the limitations of traditional diagnostic processes. For example, dental implant placement can be planned on 3D models instead of plenary images of the CBCT scans.
The current protocol can be applied in other fields of dentistry, such as endodontic microsurgery, orthotic surgery, and reconstructive surgery following facial tumor resection. Begin with the process segmentation by accessing the segment editor module, select the previously created cropped volume as the master volume of the active segmentation. Use Add to add and Remove to remove segments.
Then rename the segments according to the anatomical structure they will represent. Start the segmentation of the alveolar bone by opening the list of effects and selecting Level Tracing, a semi-automatic tool that outlines the region where pixels have the same background value as the selected pixel. Next, drag the mouse to the perimeter of the bone on one of the 2D views, and press the left mouse button to generate the segment on the selected slice of the dataset.
Then use the Paint and Erase Hand Tools to modify the segment and correct mistakes. Outline teeth and implants using the Erase tool and delete all highlighted pixels representing them. Repeat the process on every fifth slice of the dataset in the selected orientation.
Upon completing the outlining process, compute the missing segments by selecting Fill Between slices from the Effects list and click Initialize to activate contour interpolation. If the results are satisfactory, click Apply. Then scroll through the dataset upon completion to check and correct occasional mistakes.
Use the Smoothing effect by selecting Median as the smoothing method. Then set the kernel size to five by five by five pixels by adjusting the millimeter value in the bracket and clicking Apply to make segment boundaries smoother by removing protrusions. Once the segmentation of alveolar bone is completed, repeat the same steps for the segmentation of teeth.
Select segmentation from the dropdown bar to add the STL file of the intraoral scan as segmentation. Move the cursor over the module and from the sidebar, select the Fiducial Registration Wizard. From the dropdown menus in both, the From Fiducials and To Fiducial sections, choose Create New MarkupsFiducial.
In the From section next to the dropdown bar, use the Place a Markup Point icon to place marker points on well-defined anatomical landmarks on the intraoral optical scanning or IOS. Markup points will be numbered in order of placement. Place markers in the same position to create the To list.
And in the same order on the cone-beam computed tomography or CBCT dataset, markup points with the same number must represent the same anatomical landmark. After the To lists are ready, access the dropdown menu in the Registration Result Transform section of the sidebar and select Create New Linear Transform to create a transformation. Access the transforms module and select the previously created transformation as the act of transform.
In the Apply Transform section, move the IOS segmentation and the From Markups list from the Transformable box to the Transformed box. This step will help to superpose the IOS over the CBCT dataset. Open the computer-aided design or CAD software, and click Import on the home screen.
Then select the STL models previously exported from the DICOM image processing software. Go to Sculpt in the menu bar. And from the brush inventory, select the Adaptive Reduce to refine imported models.
In the sidebar, click on the Select tab and choose Brush as the selection tool. Then use Unwrap Brush mode and adjust the size of the brush. Using the brush, select the crown of each tooth until the marginal gingiva is on the IOS.
From Modify tab, select Smooth Boundary and click Apply if the results are satisfactory. Go to Select and choose Edit and Separate to create individual object from the selected area. Next, go to Analysis in the menu bar and select Inspect.
Select Flat Fill as the whole fill mode, and click Auto Repair All to create closed models from the IOS model and the separated teeth models. In the Sculpt menu, choose Shrink Smooth Brush and smooth out the edges of the filled hole. Use the Shrink Smooth Brush on the segmented tooth model until teeth are completely covered by the tooth crowns separated from the IOS.
In the object browser, select both the separated crown and the segmented model of the same tooth. In the popup sidebar, select Boolean Union and click Accept. Use Shrink Smooth to smooth the transition.
Select both bone and soft tissue models in the object browser, and then opt for the Boolean difference. Using the same process and smooth transitions as described, subtract teeth from the soft tissue model to represent the clinical situation realistically. Color the surfaces of the models by selecting Sculpt from the sidebar, and then switching the little slider from Volume to Surface.
From the brush inventory, go to Paint Vertex and use the color wheel in the Color section to select the desired color. Color the surface of each model. The representative analysis shows the outlining of the pixeled region of interest with the same background value with a yellow line.
The semi-automatic segmentation tool of level tracing was used in sagittal orientation, followed by subsequent manual segmentation. The results of the semi-automatic segmentation were refined with the help of manual tools, such as Paint and Erase. Finished segmentation was observed in axial, sagittal, and coronal view, and the 3D model was generated automatically from the segments created previously.
Super imposition of IOS and subsequent CAD modeling allowed for viewing the clinical situation in three dimensions. Level tracing is an efficient and smart edge detection tool. However, due to artifacts and scatter, the created segments may still need to be modified manually.
This process is relatively new. However, the results so far are very promising. 3D models could have a huge potential in the planning, execution, and evolution of surgical procedures in dental surgery and periodontology.
