US 12,236,594 B2
Teeth segmentation using neural networks
Sergei Azernikov, Irvine, CA (US); Fedor Chelnokov, Khimki (RU); Andrey Tolstov, Moscow (RU); and Sergey Nikolskiy, Columbia, TN (US)
Assigned to James R. Glidewell Dental Ceramics, Inc., Newport Beach, CA (US)
Filed by James R. Glidewell Dental Ceramics, Inc., Newport Beach, CA (US)
Filed on Nov. 20, 2023, as Appl. No. 18/515,179.
Application 18/515,179 is a continuation of application No. 17/140,739, filed on Jan. 4, 2021, granted, now 11,842,484.
Prior Publication US 2024/0177307 A1, May 30, 2024
Int. Cl. G06T 7/00 (2017.01); G06T 7/11 (2017.01)
CPC G06T 7/0012 (2013.01) [G06T 7/11 (2017.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/20152 (2013.01); G06T 2207/30036 (2013.01)] 9 Claims
OG exemplary drawing
 
1. A computer-implemented method of digitally segmenting teeth in a digital model, comprising:
receiving a 3D digital model of patient scan data of at least a portion of a patient's dentition;
generating a 2D depth map from the 3D digital model;
determining, using a first neural network, one or more digital tooth bounding regions in the 2D depth map;
determining one or more sampled digital surface points on the 3D digital model corresponding to one or more sampled points in the one or more digital tooth bounding regions;
generating a two dimensional (2D) panoramic image using the one or more sampled digital surface points on the 3D digital model;
labeling, using a second trained neural network, one or more regions of the 2D panoramic image to provide a labeled 2D panoramic image; and
mapping one or more regions of the labeled panoramic image to one or more corresponding coarse digital surface triangle labels in the 3D digital model to provide a labeled 3D digital model.