Machine learning scoring system and methods for tooth position assessment

Patent metadata

Field	Value
Publication number	US-11957541-B2
Application number	US-202218146327-A
Country	US
Kind code	B2
Filing date	Dec 23, 2022
Priority date	Sep 14, 2018
Publication date	Apr 16, 2024
Grant date	Apr 16, 2024

Abstract

Provided herein are systems and methods for scoring a post-treatment tooth position of a patient's teeth. A patient's dentition may be scanned and/or segmented. Raw 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 classifier can identify and/or output the post-treatment tooth position of the patient's teeth.

First claim

What is claimed is:

1. A method, comprising:
acquiring, in one or more computing devices, a digital model of a patient's teeth;
extracting, with the one or more computing devices, raw features of the patient's teeth from the digital model of the patient's teeth;
creating, in the one or more computing devices, engineered features from the raw features of the patient's teeth;
applying the raw features and/or the engineered features to a trained classifier of the one or more computing devices, wherein the trained classifier generates a post-treatment tooth position score indicating how likely a post-treatment tooth position is to be accepted or rejected by a doctor; and
outputting from the one or more computing devices the post-treatment tooth position score.

2. The method of claim 1 , wherein the digital model comprises a 3D point cloud.

3. The method of claim 2 , wherein extracting the raw features further comprises extracting up to nine points from the 3D point cloud for each of the patient's teeth.

4. The method of claim 1 , wherein extracting the raw features further comprises extracting the raw features from the digital model, from additional data about the patient or the patient's teeth, from prescription guidelines, or some combination thereof.

5. The method of claim 1 , further comprising constructing a dental appliance based on the post-treatment tooth position score.

6. The method of claim 1 , further comprising modifying the digital model based on the post-treatment tooth position score.

7. The method of claim 6 , further comprising constructing a dental appliance based on the modified digital model.

8. The method of claim 1 , further comprising processing, in the one or more computing devices, the raw features of the patient's teeth to account for one or more requirements in treatment prescriptions.

9. The method of claim 1 , further comprising processing, in the one or more computing devices, the raw features of the patient's teeth to account for missing teeth, teeth to be extracted, or teeth that are not to be moved during treatment.

10. The method of claim 1 , wherein outputting comprises outputting one or more of a binary score and a linear score indicating that the doctor is likely to reject the post-treatment tooth position or accept the post-treatment tooth position.

11. The method of claim 1 , wherein creating engineered features further comprises analyzing components of the raw features of the patient's teeth.

12. The method of claim 1 , wherein creating engineered features further comprises performing automated feature exploration of neighboring teeth dependency and local/regional arch shape from the raw features of the patient's teeth.

13. The method of claim 1 , wherein the creating engineered features step further comprises analyzing components of the raw features of the patient's teeth and performing automated feature exploration of neighboring teeth dependency and local/regional arch shape from the raw features of the patient's teeth.

14. A non-transitory computing device readable medium having instructions stored thereon for performing a computer-implemented method, wherein the instructions are executable by a processor to cause one or more computing devices to:
acquire a digital model of a patient's teeth;
extract raw features of the patient's teeth from the digital model of the patient's teeth;
create engineered features from the raw features of the patient's teeth;
apply the raw features and/or the engineered features to a trained classifier of the one or more computing devices, wherein the trained classifier generates a post-treatment tooth position score indicating how likely a post-treatment tooth position is to be accepted or rejected by a doctor; and
output the post-treatment tooth position score.

15. The non-transitory computing device readable medium of claim 14 , wherein the digital model comprises a 3D point cloud.

16. The non-transitory computing device readable medium of claim 15 , wherein the instructions are further configured to extract up to nine points from the 3D point cloud for each of the patient's teeth.

17. The non-transitory computing device readable medium of claim 14 , wherein the instructions are further configured to extract the raw features from the digital model, from additional data about the patient or the patient's teeth, from prescription guidelines, or some combination thereof.

18. The non-transitory computing device readable medium of claim 14 , wherein the instructions are further configured to modify the digital model based on the post-treatment tooth position score.

19. The non-transitory computing device readable medium of claim 14 , wherein the instructions are further configured to process the raw features of the patient's teeth to account for one or more requirements in treatment prescriptions.

20. The non-transitory computing device readable medium of claim 14 , wherein the instructions are further configured so that the output is a binary score or a linear score indicating that a doctor is likely to reject the post-treatment tooth position or accept the post-treatment tooth position.

Assignees

• Align Technology Inc

Inventors

• Li Guotu
• Cramer Christopher E
• Brown Chad Clayton
• Spiridonov Anton

Classifications

• A61C9/0053 Primary

  Optical means or methods, e.g. scanning the teeth by a laser or light beam · CPC title

• A61B5/0088

  for oral or dental tissue · CPC title

• A61B5/7264

  Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems · CPC title

• A61C7/002 Primary

  Orthodontic computer assisted systems · CPC title

• A61C13/20

  Methods or devices for soldering, casting, moulding or melting {(A61C13/04, A61C13/081 take precedence)} · CPC title