Automatic detection of tooth type and eruption status

Patent metadata

Field	Value
Publication number	US-11996181-B2
Application number	US-201816010087-A
Country	US
Kind code	B2
Filing date	Jun 15, 2018
Priority date	Jun 16, 2017
Publication date	May 28, 2024
Grant date	May 28, 2024

Abstract

Provided herein are systems and methods for detecting the eruption state (e.g., tooth type and/or eruption status) of a target tooth. A patient's dentition may be scanned and/or segmented. A target tooth may be identified. Dental features, principal component analysis (PCA) features, and/or other features may be extracted and compared to those of other teeth, such as those obtained through automated machine learning systems. A detector can identify and/or output the eruption state of the target tooth, such as whether the target tooth is a fully erupted primary tooth, a permanent partially erupted/un-erupted tooth, or a fully erupted permanent tooth.

First claim

What is claimed is:

1. A method comprising:
scanning a patient's dentition with an intraoral scanner to gather an initial three-dimensional (3D) model of the patient's dentition;
identifying a portion of the initial 3D model of the patient's dentition corresponding to a target tooth, the portion of the initial 3D model of the patient's dentition being associated with one or more visual attributes of the target tooth;
standardizing the initial 3D model of the patient's dentition by expressing at least the portion of the initial 3D model corresponding to the target tooth as vectors from a center point of the target tooth;
identifying one or more principal component analysis (PCA) features of the portion of the standardized initial 3D model corresponding to the target tooth, the one or more PCA features being correlated with the one or more visual attributes of the target tooth, wherein identifying the one or more PCA features comprises determining whether the vectors intersect a tooth shape of the target tooth and obtaining a limited number of the vectors at known angles in 3D space;
determining one or more tooth eruption indicators of the target tooth using the one or more PCA features, the one or more tooth eruption indicators providing a basis to identify an eruption state of the target tooth, wherein determining the one or more tooth eruption indicators of the target tooth comprises using a machine-trained classifier to recognize one or more of an eruption status and a tooth type of the target tooth based on one or more PCA features of a 3D model of one or more representative teeth;
predicting a spacing in the patient's dentition for accommodating eruption of the target tooth based on the one or more tooth eruption indicators;
using the predicted space to generate an orthodontic treatment plan for the patient's dentition; and
using the orthodontic treatment plan to fabricate a series of orthodontic aligners for treating the patient's dentition.

2. The method of claim 1 , wherein the one or more tooth eruption indicators comprises one of: a primary erupted tooth, a permanent partially erupted or un-erupted tooth, and a permanent erupted tooth.

3. The method of claim 1 , wherein the eruption state comprises one or more of the eruption status and a permanentness status of the target tooth.

4. The method of claim 3 , wherein the permanentness status designates whether the target tooth is a permanent tooth or a primary tooth.

5. The method of claim 1 , wherein determining the one or more tooth eruption indicators of the target tooth is based, at least in part on one or more of patient age, eruption sequence and patient gender.

6. The method of claim 1 , wherein determining the one or more tooth eruption indicators comprises comparing the one or more PCA features with the one or more PCA features of the 3D model of the one or more representative teeth.

7. The method of claim 1 , wherein determining the one or more tooth eruption indicators of the target tooth based on the PCA features comprises applying machine learning algorithms selected from the group consisting of Decision Tree, Random Forest, Logistic Regression, Support Vector Machine, AdaBOOST, K-Nearest Neighbor (KNN), Quadratic Discriminant Analysis, and Neural Network.

8. The method of claim 1 , wherein the machine-trained classifier implements a convolutional neural network.

9. The method of claim 1 , further comprising outputting a modified version of the initial 3D model of the patient's dentition to include tooth numbering based at least in part on the eruption state of the target tooth.

10. The method of claim 1 , further comprising receiving a request to identify the one or more tooth eruption indicators of the target tooth; and wherein identifying one or more principal components is performed in response to the request.

11. The method of claim 1 , wherein the orthodontic treatment plan comprises a pediatric orthodontic treatment plan for a pediatric patient.

12. The method of claim 1 , further comprising outputting the one or more tooth eruption indicators, wherein outputting the one or more tooth eruption indicators comprises providing one or more tooth eruption indicator labels corresponding to the one or more tooth eruption indicators.

13. The method of claim 1 , wherein identifying the portion of the standardized initial 3D model is part of an operation of segmenting the standardized initial 3D model of the patient's dentition.

14. The method of claim 1 , wherein the series of orthodontic aligners apply forces to the patient's teeth according to a series of treatment stages, wherein one or more of the series of treatment stages is configured to accommodate the eruption of the target tooth based on the predicted spacing.

15. The method of claim 1 , further comprising providing a visual representation of the patient's teeth including the one or more tooth eruption indicators.

16. A computer-implemented method of determining an orthodontic treatment plan for a patient's dentition, the method comprising:
converting, in a computing device, scan data from an intraoral scan of the patient's teeth into an initial three-dimensional (3D) model of the patient's teeth including a target tooth, wherein the initial 3D model is a digital mesh model;
segmenting the initial 3D model to obtain segmented tooth data representing a tooth shape of the target tooth;
determining, in the computing device, tooth shape features of the target tooth from the initial 3D model of the patient's teeth;
determining, in the computing device, tooth shape features of one or more reference teeth from the initial 3D model of the patient's teeth;
normalizing, in the computing device, at least some of the tooth shape features of the target tooth using the tooth shape features of the one or more reference teeth, wherein normalizing includes normalizing one or more of a mesial-distal width, a buccal-lingual width, a crown height, and a crown center of a 3D model of the target tooth to the one or more reference teeth;
applying a principal component analysis (PCA) on the normalized target tooth shape features to obtain one or more PCA features, wherein obtaining the PCA features comprises applying the PCA to the segmented tooth data to obtain a limited number of vectors that intersect the tooth shape of the target tooth at known angles in 3D space;
applying the PCA features to a machine-trained classifier of the computing device to determine an eruption status and an eruption permanentness status of the target tooth, wherein determining the eruption status includes determining whether the target tooth is fully erupted, partially erupted or un-erupted, wherein determining the eruption status and the eruption permanentness status includes determining whether the target tooth is a primary tooth or a permanent tooth;
predicting a spacing in the patient's dentition for accommodating eruption of the target tooth based on the eruption status and the eruption permanentness status;
using the predicted space to generate the orthodontic treatment plan for the patient's dentition; and
using the orthodontic treatment plan to fabricate a series of orthodontic aligners for treating the patient's dentition.

17. The computer-implemented method of claim 16 , wherein normalizing includes determining a total number of cusps in buccal-mesial, buccal-distal, lingual-mesial, and lingual-distal arch direction surfaces.

18. The computer-implemented method of claim 16 , further comprising collecting patient information including one or more of the pat

Assignees

• Align Technology Inc

Inventors

• Xue Ya
• Choi Jeeyoung
• Moore Justin B
• Spiridonov Anton

Classifications

• G06N3/09

  Supervised learning · CPC title

• G06N3/0464

  Convolutional networks [CNN, ConvNet] · CPC title

• G16H30/40 Primary

  for processing medical images, e.g. editing · CPC title

• A61C7/002 Primary

  Orthodontic computer assisted systems · CPC title

• A61C7/08

  Mouthpiece-type retainers {or positioners, e.g. for both the lower and upper arch (arch-shaped medicament applicators for teeth or gums A61C19/063; mouthguards for protecting the teeth of sportsmen A63B71/085)} · CPC title