Systems and methods for segmenting dental scans
Systems and methods for segmenting dental scans. First and second segmented volumes of scan data may be formed, where the first segmented volume may have a lower resolution than the segmented volume. Segmenting scan data may be formed by merging the first segmented volume with the second segmented volume. Merging may include removing teeth labels from the first segmented volume, adjusting a resolution of the first segmented volume to be the same as a resolution of the second segmented volume, and replacing voxels in the first segmented volume with voxels from the second segmented volume. The segmented scan data may be aligned to a digital model of the patient's dentition. The digital model of the patient's dentition may be modified to include the tooth roots from the aligned segmented scan data to form a modified digital model. A treatment plan may be displayed or modified using the modified digital model.
1 . A method, the method comprising:
receiving, in a computing device, scan data comprising a bone root scan of a patient's dentition;
segmenting the scan data into individual teeth, wherein the individual teeth include crowns and tooth roots, wherein the segmenting comprises:
forming a first segmented volume by segmenting a first portion of the scan data with a first segmentation process;
forming a second segmented volume by segmenting a second portion of the scan data with a second segmentation process, wherein the first segmented volume has a lower resolution than the second segmented volumes; and
forming segmenting scan data by merging the first segmented volume with the second segmented volume, the merging comprising:
removing teeth labels from the first segmented volume;
adjusting a resolution of the first segmented volume to be the same as a resolution of the second segmented volume; and
replacing voxels in the first segmented volume with voxels from the second segmented volume;
aligning the segmented scan data to a digital model of the patient's dentition;
modifying the digital model of the patient's dentition to include the tooth roots from the aligned segmented scan data to form a modified digital model; and
displaying or modifying a treatment plan using the modified digital model.
2 . The method of claim 1 , wherein the bone root scan comprises a cone beam computed tomography (CBCT) scan.
3 . The method of claim 1 , wherein the bone root scan comprises one or more of: a cone beam computed tomography (CBCT) scan, a CT scan, and an MRI scan.
4 . The method of claim 1 , wherein segmenting the scan data comprises using a first neural network to segment the first portion of the scan data and a second neural network to segment the second portion of the scan data.
5 . The method of claim 1 , further comprising processing the aligned segmented scan data before modifying the digital model, wherein processing the aligned segmented scan data comprises patching teeth sockets from the aligned segmented scan data, wherein the teeth sockets correspond to parts of bone of the scan data where the individual teeth exist.
6 . The method of claim 1 , wherein the aligning comprises performing a coarse alignment and a fine alignment, wherein performing the fine alignment comprises applying an Iterative Closest Point (ICP), 3D matching algorithm.
7 . The method of claim 1 , further comprising smoothing the aligned segmented scan data.
8 . The method of claim 1 , further comprising processing the aligned segmented scan data to patch teeth sockets from the aligned segmented scan data, wherein patching the teeth sockets comprises removing tooth sockets, and displaying the modified digital model so that semitransparent representations of the bone are displayed over planned tooth movement without visual interference of moving root contours and unmovable socket contours.
9 . The method of claim 1 , wherein the aligning comprises a coarse alignment and a fine alignment, wherein the coarse alignment comprises determining the most prominent tip point of the crowns of each of the individual teeth using vectors from centers of the individual teeth to centers of teeth in an opposite jaw of the patient, and aligning the most prominent tip points of the crowns of the segmented scan data with corresponding tip points of the crowns of the digital model.
10 . The method of claim 9 , further comprising stitching the crowns of the digital model to corresponding tooth roots of the aligned segmented scan data.
11 . The method of claim 10 , wherein stitching the crowns of the digital model to the corresponding tooth roots of the aligned segmented scan data comprises straightening a stitching line of the aligned segmented scan data using a straightening transform.
12 . The method of claim 1 , wherein the first segmentation process is a binary segmentation process that categorizes pixels into two classes, wherein the second segmentation process is a multi-class segmentation process that categorizes pixels into more than two classes.
13 . The method of claim 1 , wherein the first segmentation volume comprises information associated with bones and the second segmentation volume comprises information associated with teeth.
14 . A dental treatment system comprising:
one or more processors;
memory coupled to the one or more processors, the memory configured to store computer-program instructions that, when executed by the one or more processors, cause the one or more processors to execute a computer-implemented method, the computer-implemented method comprising:
receiving, in a computing device, scan data comprising a bone root scan of a patient's dentition;
segmenting the scan data into individual teeth, wherein the individual teeth include crowns and tooth roots, wherein the segmenting comprises:
forming a first segmented volume by segmenting a first portion of the scan data with a first segmentation process;
forming a second segmented volume by segmenting a second portion of the scan data with a second segmentation process, wherein the first segmented volume has a lower resolution than the second segmented volumes; and
forming segmenting scan data by merging the first segmented volume with the second segmented volume, the merging comprising:
removing teeth labels from the first segmented volume;
adjusting a resolution of the first segmented volume to be the same as a resolution of the second segmented volume; and
replacing voxels in the first segmented volume with voxels from the second segmented volume;
aligning the segmented scan data to a digital model of the patient's dentition;
modifying the digital model of the patient's dentition to include the tooth roots from the aligned segmented scan data to form a modified digital model; and
displaying or modifying a treatment plan using the modified digital model.
15 . The dental treatment system of claim 14 , wherein the scan data is three-dimensional (3D) scan data and the digital model is a 3D digital model.
16 . The dental treatment system of claim 14 , wherein the aligning comprises determining the most prominent tip point of the crowns of each of the individual teeth using vectors from centers of the individual teeth to centers of teeth in an opposite jaw of the patient, and aligning the most prominent tip points of the crowns of the segmented scan data with corresponding tip points of the crowns of the digital model.
17 . The dental treatment system of claim 16 , wherein the computer-implemented method further comprises stitching the crowns of the digital model to corresponding tooth roots of the aligned segmented scan data.
18 . The dental treatment system of claim 17 , wherein stitching the crowns of the digital model to the corresponding tooth roots of the aligned segmented scan data comprises straightening a stitching line of the scan data using a straightening transform.
19 . The dental treatment system of claim 14 , further comprising displaying the modified digital model with semitransparent representations of the bone over planned tooth movement without visual interference of moving root contours and unmovable socket contours.
20 . The dental treatment system of claim 14 , wherein the computer-implemented method further comprises processing the aligned segmented scan data before modifying the digital model, wherein processing the aligned segmented scan data comprises patching teeth sockets from the aligned segmented scan data, wherein the teeth sockets correspond to parts of bone of the scan data where the individual teeth exist.
21 . The dental treatment system of claim 20 , wherein the computer-implemented method further comprises displaying raw aligned segmented scan data for a user to review and correct the modified digital model.
22 . The dental treatment system of claim 14 , wherein the aligning comprises performing a coarse alignment and a fine alignment, wherein performing the fine alignment comprises applying an Iterative Closest Point (ICP), 3D matching algorithm.
23 . The dental treatment system of claim 14 , wherein the bone root scan comprises a cone beam computed tomography (CBCT) scan.
24 . The dental treatment system of claim 14 , wherein the bone root scan comprises a CT scan.
25 . The dental treatment system of claim 14 , wherein the bone root scan comprises an MRI scan.
26 . The dental treatment system of claim 14 , wherein segmenting the scan data comprises using a first neural network to segment the first portion of the scan data and a second neural network to segment the second portion of the scan data.
27 . The dental treatment system of claim 14 , wherein the first segmentation process is a binary segmentation process that categorizes pixels into two classes, wherein the second segmentation process is a multi-class segmentation process that categorizes pixels into more than two classes.
28 . The dental treatment system of claim 14 , wherein the first segmentation volume comprises information associated with bones and the second segmentation volume comprises information associated with teeth.