I’ve created a comprehensive video tutorial that walks through my full workflow for performing virtual surgical planning (VSP) for orthognathic surgery.
The video covers everything from aligning CT scans and intra-oral scan data to performing virtual osteotomies and designing the final surgical splint. I’ve focused on using only 3D Slicer.
I hope this resource is helpful for other surgeons, residents, and anyone interested in advancing their digital planning skills. Feedback and discussion are welcome!
Thank you so much for interesting work. How can we accurately register an intraoral scan to teeth extracted from CBCT? We used FastModelAlign to align the intraoral scan with the CBCT tooth segmentation. However, the registration evaluation reported a distance of approximately 10.30 mm, or about 1 cm. This error is unacceptable for real time application because it could create significant positional errors in the complete 3D structure. Nevertheless, we should not conclude that FastModelAlign itself caused geometric distortion until we confirm which distance metric was measured and whether both input models contained the same anatomical regions. How can we validate the registration result and determine whether the transform generated by FastModelAlign is reliable?
Hi Sorry for the delayed response. Where was it reported as 10.30 mm error ? In any case you might have have to optimize the scan and the segments to get the best results (trimming unwanted parts) and then applying the transformation to the full scan. My experience is that you can get the same results as with ExoPlan or Materialize Enlight or mimics with much better control of the segmentations. As in any other software you should verify manually the alignment by visual inspection against the CT/CBCT and the algner IOS. Also you can use heatmap (model to Model distance module) .
We used segment comparison model. After trimming a result slightly affected. Tried all parameters from Fast Align tool. This error is unacceptable for real time application because it could create significant positional errors in the complete 3D structure.
Did you check against the CT visually ? do you see 1 Cm deviation ? i am not sure how comparison model module works ? Visually see how the the model is aligned. Also as control test export the aligned IOS and the segment from Exoplan or Materialize enlight see the results with the comparison model module ? in any case if you visually inspect aligned model vs CT you might be able to better see what has happened.
