Intraoral imaging before and after treatment

1 . A method comprising:

acquiring a reference 3D model of a patient's jaw before a procedure commences;

after the procedure is completed and responsive to implementing a modification to an area of interest of the patient's jaw, acquiring an intraoral scan of the area of interest to generate an after-treatment 3D model; and

unifying the reference 3D model with the after-treatment 3D model using a blending by distance operation performed on transition areas of the reference 3D model and the after-treatment 3D model, wherein the after-treatment 3D model is representative of only a portion of the reference 3D model.

2 . The method of claim 1 , wherein unifying the reference 3D model with the after-treatment 3D model comprises:

establishing a first coordinate system for the reference 3D model and a second coordinate system for the after-treatment 3D model;

aligning the first coordinate system with the second coordinate system to overlay the after-treatment 3D model on the reference 3D model, and

reconstructing an overall 3D model by replacing at least some portions of the reference 3D model corresponding to the after-treatment 3D model with the after-treatment 3D model.

3 . The method of claim 2 , wherein changed areas corresponding to the modification are identified between the first coordinate system and the second coordinate system.

4 . The method of claim 2 , wherein aligning the first coordinate system with the second coordinate system comprises performing a fast point feature histogram (FPFH) method.

5 . The method of claim 4 , wherein if non-contiguous portions of the patient's jaw are recorded in the after-treatment 3D model, pre-registration of each non-contiguous portion is performed.

6 . The method of claim 2 , wherein responsive to the aligning, a global compensation is performed to correct a deviation between the reference 3D model and the after-treatment 3D model by registering individual images of the after-treatment 3D model against the reference 3D model without using correspondences between the individual images and the reference 3D model at areas corresponding to the modification.

7 . The method of claim 6 , wherein the global compensation is performed for new individual images of the after-treatment 3D model in real time without using said correspondences that are based on the new individual images.

8 . The method of claim 6 , wherein:

individual images corresponding to the modification are determined by computing an error between the reference 3D model and individual images of the after-treatment 3D model, and

responsive to computing that the error exceeds an error threshold, said individual images of the after-treatment 3D model are marked as individual images corresponding to the modification and ignored.

9 . The method of claim 2 , wherein the reconstructing further comprises:

computing, for an implicit representation of the reference 3D model, an overlapping area of the after-treatment 3D model;

computing a plurality of areas inside the overlapping area for use in the blending by distance operation, each area of the plurality of areas being assigned a corresponding predetermined weight;

computing, for an implicit representation of the after-treatment 3D model, a plurality of other areas corresponding to the plurality of areas, for use in the blending by distance operation, each area of the plurality of other areas being assigned another corresponding predetermined weight that is complementary to the corresponding predetermined weight; and

converting the implicit representation of the reference 3D model and the implicit representation of the after-treatment 3D model into the overall 3D model using the blending by distance operation wherein areas of the reference 3D model and the after-treatment 3D model contribute to the overall 3D model according to the corresponding predetermined weights of the plurality of areas and the another corresponding predetermined weights of the another plurality of areas, respectively.

10 . The method of claim 9 , further comprising:

converting the implicit representation of the reference 3D model and the implicit representation of the after-treatment 3D model into the overall 3D model using marching cubes.

11 . The method of claim 1 , wherein the modification is a scan body or a preparation of a removed gingival holder.

12 . A system comprising:

memory to store instructions; and

a processor configured by the instructions to:

acquire a reference 3D model of a patient's jaw before a procedure commences;

after the procedure is completed and responsive to implementing a modification to an area of interest of the patient's jaw, acquire an intraoral scan of the area of interest to generate an after-treatment 3D model; and

unify the reference 3D model with the after-treatment 3D model using a blending by distance operation performed on transition areas of the reference 3D model and the after-treatment 3D model, wherein the after-treatment 3D model is representative of only a portion of the reference 3D model.

13 . The system of claim 12 , wherein to unify the reference 3D model with the after-treatment 3D model the processor is further configured to:

establish a first coordinate system for the reference 3D model and a second coordinate system for the after-treatment 3D model;

align the first coordinate system with the second coordinate system to overlay the after-treatment 3D model on the reference 3D model, and

reconstruct an overall 3D model by replacing at least some portions of the reference 3D model corresponding to the after-treatment 3D model with the after-treatment 3D model.

14 . The system of claim 13 , wherein the processor is further configured to:

align the first coordinate system with the second coordinate system by performing a fast point feature histogram (FPFH) method.

15 . The system of claim 14 , wherein if non-contiguous portions of the patient's jaw are recorded in the after-treatment 3D model, pre-registration of each non-contiguous portion is performed.

16 . The system of claim 13 , wherein the processor is further configured to:

perform, responsive to the aligning, a global compensation to correct a deviation between the reference 3D model and the after-treatment 3D model by registering individual images of the after-treatment 3D model against the reference 3D model without using correspondences between the individual images and the reference 3D model at areas corresponding to the modification.

17 . The system of claim 16 , wherein the processor is further configured to perform the global compensation for new individual images of the after-treatment 3D model in real time without said correspondences that are based on the new individual images.

18 . The system of claim 16 , wherein the processor is further configured to:

determine individual images corresponding to the modification by computing an error between the reference 3D model and individual images of the after-treatment 3D model, and

mark, responsive to computing that the error exceeds an error threshold, said individual images of the after-treatment 3D model as individual images corresponding to the modification and ignored.

19 . The system of claim 13 , wherein to reconstruct, the processor is further configured to:

compute, for an implicit representation of the reference 3D model, an overlapping area of the after-treatment 3D model;

compute a plurality of areas inside the overlapping area for use in the blending by distance operation, each area of the plurality of areas being assigned a corresponding predetermined weight;

compute, for an implicit representation of the after-treatment 3D model, a plurality of other areas corresponding to the plurality of areas, for use in the blending by distance operation, each area of the plurality of other areas being assigned another corresponding predetermined weight that is complementary to the corresponding predetermined weight; and

convert the implicit representation of the reference 3D model and the implicit representation of the after-treatment 3D model into the overall 3D model using the blending by distance operation wherein areas of the reference 3D model and the after-treatment 3D model contribute to the overall 3D model according to the corresponding predetermined weights of the plurality of areas and the another corresponding predetermined weights of the another plurality of areas, respectively.

20 . A non-transitory computer-readable storage medium including instructions that when executed by a computer, cause the computer to:

acquire a reference 3D model of a patient's jaw before a procedure commences;

after the procedure is completed and responsive to implementing a modification to an area of interest of the patient's jaw, acquire an intraoral scan of the area of interest to generate an after-treatment 3D model; and

unify the reference 3D model with the after-treatment 3D model using a blending by distance operation performed on transition areas of the reference 3D model and the after-treatment 3D model, wherein the after-treatment 3D model is representative of only a portion of the reference 3D model.